Final Revised Willamette Basin Mercury Total Maximum Daily Load

Nov. 22, 2019

Watershed Management 700 NE Multnomah St. Suite 600 Portland, OR 97232 Phone: 503-229-5696 800-452-4011 Fax: 503-229-6762 Contact: Alex Liverman, Andrea Matzke, Kevin Brannan, Priscilla Woolverton www.oregon.gov/DEQ

DEQ is a leader in restoring, maintaining and enhancing the quality of Oregon’s air, land and water.

This report prepared by:

Oregon Department of Environmental Quality 700 NE Multnomah Street Portland, OR 97232 1-800-452-4011 www.oregon.gov/deq

Contact: Kevin Brannan Alex Liverman Andrea Matzke Priscilla Woolverton 503-229-6629 503-229-5080 503-229-5350 541-687-7347

Alternative formats: DEQ can provide documents in an alternate format or in a language other than English upon request. Call DEQ at 800-452-4011 or email [email protected].

Oregon Department of Environmental Quality ii Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Table of Contents

Executive Summary ...... 1 1. TMDL Introduction ...... 3 1.1 TMDL Authority ...... 3 1.2 General TMDL approach...... 3 1.3 History of Willamette Basin Mercury TMDL ...... 6 1.4 Name and location ...... 7 1.4.1 HUCs, subbasins, and watersheds ...... 7 1.4.2 Land use ...... 9 1.4.3 Climate ...... 11 1.4.4 Stream Flow ...... 11 2. Beneficial uses ...... 12 3. Applicable water quality standards ...... 13 4. Mercury water quality impairments ...... 15 5. Summary of Mercury TMDL development and approach ...... 17 5.1 Mercury cycling in the environment ...... 17 5.2 Mercury TMDL approach ...... 18 6. Explanation of models and current mercury load ...... 20 6.1 Explanation of models ...... 20 6.1.1 Food Web Model ...... 21 6.1.2 Mercury translator equation ...... 22 6.1.3 Linkage analysis using Mass Balance Model ...... 25 6.1.4 Nonpoint source input data development ...... 30 6.1.4.1 Atmospheric deposition ...... 30 6.1.4.2 Soil ...... 30 6.1.4.3 Groundwater ...... 31 6.1.5 Estimation of instream delivery of mercury and reservoir processes ...... 31 6.2 Current total mercury load estimation for the Willamette Basin ...... 32 7. Loading capacity and excess load ...... 34 7.1 Loading capacity ...... 34 7.2 Excess load ...... 34 8. Seasonal variation and critical condition ...... 39 9. Source assessment ...... 40 9.1 Air emissions ...... 40

State of Oregon Department of Environmental Quality iii Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

9.1.1 Oregon permitted mercury air emissions ...... 40 9.1.2 National and global mercury emissions trends ...... 40 9.2 Nonpoint sources ...... 41 9.2.1 Agriculture, forestry, non-MS4 permitted urban areas, and water conveyances ...41 9.2.2 Impoundments ...... 41 9.2.3 Legacy metals mining sources ...... 44 9.3 Background and unquantified anthropogenic sources ...... 46 9.4 Point sources ...... 47 9.4.1 Wastewater permits and mercury loads ...... 47 9.4.1.1 Municipal sewage treatment plant permits ...... 47 9.4.1.2 Industrial wastewater permits ...... 49 9.4.2 Stormwater permits and mercury loads ...... 51 9.4.2.1 Municipal stormwater permits ...... 52 9.4.2.2 Industrial Stormwater General Permits ...... 54 9.4.2.3 Construction Stormwater Permits...... 54 10. Allocations ...... 55 10.1 Load allocations for nonpoint sources ...... 58 10.2 Wasteload allocations for point sources ...... 59 10.3 Instream surrogate targets ...... 63 11. Margin of safety ...... 65 12. Reserve capacity ...... 66 13. Water Quality Management Plan ...... 67 13.1 Introduction ...... 67 13.1.1 Implementation plans ...... 68 13.1.2 Adaptive management ...... 69 13.2 Elements of the Water Quality Management Plan ...... 71 13.2.1 Condition assessment and problem description ...... 71 13.2.2 Goals and objectives ...... 72 13.2.3 Identification of designated management agencies and responsible persons ...... 72 13.3 Management strategies ...... 73 13.3.1 Management strategies for nonpoint sources and water protection programs ...... 74 13.3.1.1 Oregon Department of Environmental Quality Nonpoint Source ...... 75 13.3.1.2 DEQ Cleanup Program—Abandoned Mine Lands Sites ...... 76 13.3.1.3 DEQ Cleanup Program—Portland Harbor Superfund Source Control ...... 76 13.3.1.4 Oregon Department of Agriculture ...... 77 13.3.1.5 Oregon Department of Forestry ...... 81 13.3.1.6 Oregon Department of State Lands ...... 84

State of Oregon Department of Environmental Quality iv Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

13.3.1.7 Oregon Parks and Recreation Department ...... 87 13.3.1.8 Oregon Department of Geology and Mineral Industries ...... 88 13.3.1.9 Oregon Department of Fish and Wildlife ...... 88 13.3.1.10 Oregon State Marine Board ...... 89 13.3.1.11 Cities ...... 90 13.3.1.12 Counties ...... 96 13.3.1.13 Bureau of Land Management ...... 102 13.3.1.14 U.S. Forest Service ...... 104 13.3.1.15 U.S. Fish and Wildlife Service ...... 106 13.3.1.16 U.S. Environmental Protection Agency ...... 107 13.3.1.17 Metro (Portland Metropolitan Government) ...... 107 13.3.1.18 Port of Portland ...... 108 13.3.1.19 Clean Water Services ...... 108 13.3.1.20 Tualatin Hills Park and Recreation District ...... 109 13.3.1.21 Oak Lodge Water Services District ...... 109 13.3.1.22 Water Environment Services ...... 110 13.3.1.23 Water Delivery and Conveyance Systems ...... 110 13.3.1.24 Reservoir management ...... 114 13.3.2 Management strategies for point sources ...... 117 13.3.2.1 NPDES Wastewater Permits ...... 117 13.3.2.2 NPDES Permitted Stormwater ...... 120 13.3.3 Other DEQ Mercury Reduction Programs ...... 122 13.3.3.1 Regulatory Programs ...... 122 13.3.3.2 Voluntary programs ...... 123 13.4 Timeline for implementing management strategies ...... 124 13.4.1 Nonpoint Source DMAs and responsible persons ...... 125 13.4.2 Point sources ...... 126 13.5 Timeline for attainment of water quality standards ...... 127 13.6 Monitoring and evaluation ...... 127 13.7 Costs and funding ...... 128 13.8 Citation legal authorities ...... 131 14. Reasonable Assurance ...... 135 14.1 Accountability framework ...... 136 14.1.1 Pollutant reduction strategies ...... 136 14.1.2 Identify relevant DMAs ...... 137 14.1.3 Develop timeline, targets, measurable objectives ...... 137 14.1.4 Evaluate implementation plans and progress ...... 137 14.1.5 Take action on failure to implement ...... 139 14.1.6 Track water quality status and trends ...... 139

State of Oregon Department of Environmental Quality v Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

14.2 Dominance of atmospheric deposition of mercury ...... 139 14.3 Conclusions ...... 140 15. References ...... 141 Appendix A. Technical Support Document ...... 147 Appendix B. Calculation of allocations ...... 148 Appendix C. Variance justification excerpts ...... 150 Appendix D. Stormwater references and resources ...... 157 Appendix E. List of designated management agencies and responsible persons ...... 158 Appendix F. Oregon permitted mercury air emissions ...... 167 Appendix G. Draft of Willamette River Instream Surrogate TSS-THg Analysis ...... 171 1.1 Introduction ...... 172 1.2 Methods ...... 172 1.2.1 Addressing Non-Detect (Censored) data ...... 173 1.2.2 Ordinary Least Squares ...... 174 1.2.3 Linear Mixed-Effects Model ...... 174 1.2.4 Prediction Intervals ...... 175 1.3 Results ...... 175 1.4 Discussion ...... 176 1.5 Recommendations ...... 176 1.6 Tables ...... 178 1.7 Figures ...... 187 2. References ...... 200 3. R Packages Used ...... 201 Appendix H. Watershed-Based Plan Crosswalk ...... 202 Appendix I. Select HUC8-Level Analysis of Existing Condition and Conservatism of At- Source vs. Delivered Loads ...... 206

State of Oregon Department of Environmental Quality vi Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

List of Tables

Table 1-1. Summary of Willamette Basin Mercury TMDL Components ...... 4 Table 1-2. HUC8 codes and corresponding watershed names in the Willamette Basin...... 7 Table 1-3. Land Use Areas and Percentages in Willamette Basin...... 9 Table 4-1. 303(d) Listings for Mercury in the Willamette Basin ...... 15 Table 6-1. Fish Species and Associated Trophic Levels used in the Food Web Model ...... 21 Table 6-2. Biomagnification (L/kg) estimates from Food Web Model for fish species...... 22 Table 6-3. Slope estimates and associated statistics for translator equation...... 23 Table 6-4. Species-specific Surface Water THg Target Levels to Meet a Fish Tissue Concentration of 0.040 mg/kg MeHg...... 25 Table 6-5. Estimated Wet and Dry Atmospheric Deposition Rates in the Willamette Basin...... 30 Table 6-6. Soil total mercury potency factor estimates for Willamette Basin ...... 31 Table 6-7. Estimate total mercury loads for source categories from Mass Balance Model for Willamette Basin...... 32 Table 7-1. Required reductions of the existing median surface water total mercury concentration of 1.2 ng/L to meet fish species specific surface water total mercury target levels...... 38 Table 9-1. Reservoirs Represented in the Willamette Basin HSPF Model...... 44 Table 9-2. Abandoned Mine Lands and Mining Districts Within the Willamette Basin...... 45 Table 9-3. Summary of Major NPDES-Permitted Municipal Sewage Treatment Facilities in the Willamette Basin...... 48 Table 9-4. Summary of Individual NPDES-Permitted Industrial Facilities in the Willamette Basin with Activities that could Increase Mercury in their Discharge...... 49 Table 9-5. Summary of NPDES MS4 Phase I and Phase II Jurisdictions in the Willamette Basin...... 52 Table 10-1. Summary of TMDL Components...... 56 Table 10-2. Schedule for instream mainstem Willamette and HUC8 outlets surrogate targets for reducing high total suspended solid concentrations during times of high precipitation events and high flows...... 64 Table 13-1. TMDL implementation reports and summaries ...... 74 Table 13-2. Summary of DEQ programs that have the potential to reduce mercury loading in the Willamette Basin...... 75 Table 13-3. Example list of management strategies that are currently included in Agricultural Water Quality Management Area Plans to address sediment and erosion...... 79 Table 13-4. ODF Rules Related to Water Quality and Erosion Control ...... 81 Table 13-5. Example list of management strategies that ODF currently implements to address sediment and erosion...... 82 Table 13-6. Management Strategies that Department of State Lands currently requires of people that apply for, and hold authorizations to use, state-owned land, which are protective of water quality in the Willamette Basin...... 85 Table 13-7. Example list of management strategies to reduce sediment and erosion...... 86 Table 13-8. Example list of management strategies that OPRD currently implements to address sediment and erosion...... 88 Table 13-9. Example list of management strategies that ODFW currently implements to address sediment and erosion...... 89 Table 13-10. Stormwater Summary Statistics ...... 91 Table 13-11. Minimum requirements for cities...... 92 Table 13-12. Minimum management program requirements for counties...... 97 Table 13-13. Example List of County Management Strategies to Reduce Sediment/Mercury ...98 Table 13-14. Stormwater Control Measures Implementation Schedule for Cities...... 100

State of Oregon Department of Environmental Quality vii Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Table 13-15. Management Program Implementation Schedule for Counties...... 101 Table 13-16. Example list of management strategies that BLM currently implements to reduce sediment and erosion...... 103 Table 13-17. Example list of management strategies that USFS currently implements to reduce sediment and erosion...... 105 Table 13-18. Example list of management strategies that USFWS currently implements to reduce sediment and erosion...... 106 Table 13-19. Example list of management strategies that THPRD currently implements to reduce sediment and erosion...... 109 Table 13-20. Existing state and federal agencies and programs that regulate water conveyance systems...... 111 Table 13-21. Milestones and timelines for DEQ to work with water conveyance entities to plan and carry out implementation of the 2019 Willamette Basin Mercury TMDL ...... 112 Table 13-22. Examples of Management Strategies that may be required of water conveyance entities named as responsible persons in the 2019 Willamette Basin Mercury TMDL WQMP . 113 Table 13-23. Example of Best Management Practices for Reservoirs ...... 116 Table 13-24. The timeline for activities related to this WQMP and associated DMA and responsible person Implementation plans, and NPDES permits...... 125 Table 13-25. Timeline for reaching interim milestones for the general nonpoint source 88 percent reduction in instream mercury levels...... 127 Table 13-26. Partial list of funding programs available in the Willamette Basin that may be used to support planning and implementation activities that benefit water quality ...... 129 Table B-1. Source category names and descriptions used in spreadsheet for calculating the allocations...... 148 Table B-2. Allocation spreadsheet table...... 149 Table B-3. Comparison of Load Capacity to TMDL Loads Calculated for the Reductions...... 149 Table C-1. Potential treatment technologies considered for mercury treatment ...... 152 Table C-2. Treatment capability of mercury technologies ...... 153 Table F-1. 2016 Reported Mercury Emissions from Permitted Sources in the Willamette Basin ...... 167 Table G-1. Full TSS-THg Dataset Provided by TetraTech for the Instream Willamette River Basin Surrogate Analysis...... 178 Table G-2. Sampling Sites for the Instream Willamette River TSS-THg Surrogate Analysis .... 183 Table G-3. Summary Statistics for Untransformed TSS (mg/L) and THg Concentrations (ng/L) across all HUC 8 Subbasins ...... 183 Table G-4. Summary Statistics for Untransformed THg Concentrations (mg/L) for each individual HUC 8 Subbasin...... 183 Table G-5. Summary Statistics for Untransformed TSS Concentrations (ng/L) for each individual HUC 8 Subbasin ...... 184 Table G-6. OLS Model Summary Results for all HUC 8 Subbasins...... 184 Table G-7. LME Model Summary Results for all HUC 8 Subbasins with Sites as a Random Effect and Dry/Wet Seasons as a Fixed Effect...... 184 Table G-8. Values for adjusting the intercept in the LME model for total mercury and TSS with season and site as random effects...... 184 Table G-9. LME Model Summary Results for all HUC 8 Subbasins with Sites as a Random Effect but without the Fixed Effect of Dry/Wet Seasons...... 185 Table G-10. Values for adjusting the intercept in the LME model for total mercury and TSS with site as a random effect...... 185 Table G-11. Estimated Concentrations of TSS based on the LME Model Equation ...... 185 Table G-12. Estimated Percent Reduction of THg Concentrations based on the LME Model .. 186 Table G-13. Schedule for interim goals for reducing TSS concentrations ...... 186

State of Oregon Department of Environmental Quality viii Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Table H-1. Watershed Based Plan Checklist for the Willamette Basin Mercury TMDL ...... 202 Table I-1. General characteristics of recent total mercury data...... 206 Table I-2. Annual Average sediment associated loads for Dairy Creek ...... 208

List of Figures

Figure 1-1.Willamette Basin ...... 8 Figure 1-2. Land use distribution in Willamette Basin. Adapted from figure in Appendix: Technical Support Document...... 10 Figure 4-1. Mercury Impairments in the Willamette Basin ...... 16 Figure 5-1. Linked models and forms of mercury used to develop the Willamette Basin Mercury TMDL...... 19 Figure 6-1. Conceptual model for estimating load capacity from fish-tissue MeHg to water column THg...... 20 Figure 6-2. Comparison of Food Web Model biomagnification estimates to national values bioaccumulation factors values...... 22 Figure 6-3. Final Mercury Translator Model: Aggregated, Year-round, Zero-Intercept Model by HUC8 Weighted by Sample Size...... 24 Figure 6-4. Conceptual model of linking total mercury source in the watershed to total mercury in the streams and rivers of the Willamette Basin...... 26 Figure 6-5. Conceptual Framework for the Total Mercury Mass Balance Model. Figure taken from Appendix: Technical Support Document...... 27 Figure 6-6. Existing HSPF Model Domain for the Willamette Basin. Taken from Appendix: Technical Support Document...... 28 Figure 6-7. Schematic of Model Hydrologic Response Unit (HRU) Development. Taken Appendix: Technical Support Document...... 29 Figure 7-1. Boxplots of observed total mercury concentrations by HUC8 (See Table 1-2) and medians of entire data sets with data from Coast Fork and without...... 35 Figure 7-2. Comparison of the individual medians of the HUC08 total mercury data to the distribution of the data represented by the boxplot of the data used to calculate the basin-wide existing condition of 1.2 ng/L...... 36 Figure 7-3. Comparison of the individual total mercury medians of the Lower Willamette and Tualatin HUC08s using total mercury data collected in the mainstems between 2012 and 2019 to the distribution of the data represented by the boxplot of the data used to calculate the basin- wide existing condition of 1.2 ng/L...... 37 Figure 9-1. U.S. Army Corps of Engineers Dams and Reservoirs in the Willamette Basin...... 43 Figure 9-2. PGE’s North Fork Dam and Reservoir on the Clackamas River...... 44 Figure 9-3. Monthly THg loads from Cottage Grove Reservoir...... 46 Figure 13-1. McKenzie River ...... 67 Figure 13-2. Conceptual representation of adaptive management. The estimated timeline for achieving water quality standards is multiple decades...... 70 Figure 13-3. Number of Designated Management Agencies and Responsible Persons Identified in the Willamette Basin Mercury TMDL ...... 73 Figure 13-4. Middle Fork Willamette River...... 87 Figure 13-5. Map of Reservoirs Belonging to the Four Largest Owners in the Willamette Basin ...... 115 Figure 13-6. Multnomah Channel, Lower Willamette...... 128 Figure 14-1. A Representation of the Reasonable Assurance Accountability Framework Led by DEQ...... 136

State of Oregon Department of Environmental Quality ix Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Figure C-1. Number of Wisconsin municipal wastewater treatment systems with increasing and decreasing trends in average (left) and 4-day P99 (right) concentrations. (Wisconsin DNR). .. 153 Figure C-2. Number of Wisconsin municipal WWTPs by 4-day P99 mercury concentrations from initial five-year period (left) to most recent five-year period (right)...... 154 Figure C-3. Number of Wisconsin industrial wastewater treatment systems with increasing and decreasing trends in average (left) and 4-day P99 (right) concentrations...... 154 Figure C-4. Number of Wisconsin industrial NPDES facilities by 4-day P99 mercury concentrations from initial five-year period (left) to most recent five-year period (right)...... 155 Figure C-5. Influent Data from Major Wastewater Treatment Plants in Minnesota. Source: Minnesota Pollution Control Agency ...... 155 Figure C-6. Mercury Concentrations in Biosolids, Rock Creek Wastewater Treatment Plan. Source: Clean Water Services...... 156 Figure G-1. Boxplot of Untransformed TSS (mg/L) and THg Concentrations (ng/L) from the Original Dataset of 187 Surrogate Pairs. The Red line represents an Instream Target Mercury Concentration of 0.14 ng/L...... 187 Figure G-2. Dry and Wet Season Scatterplot of Untransformed TSS (mg/L) and THg Concentrations (ng/L) from the Original Dataset of 187 Surrogate Pairs...... 188 Figure G-3. Willamette River Basin Instream Surrogate TSS-THg Sampling Sites Map...... 189 Figure G-4. Boxplot of Untransformed TSS (mg/L) and THg Concentrations (ng/L) based on the Dataset of 63 Surrogate Pairs...... 190 Figure G-5. Scatterplot of Untransformed TSS (mg/L) and THg Concentrations (ng/L) according to HUC 8 Subbasin based on the Dataset of 63 Surrogate Pairs...... 191 Figure G-6. Scatterplot of TSS (mg/L) and THg Concentrations (ng/L) for each individual HUC 8 Subbasin based on the Dataset of 63 Surrogate Pairs...... 191 Figure G-7. Dry and Wet Season Scatterplot of Untransformed TSS (mg/L) and THg Concentrations (ng/L) based on the Dataset of 63 Surrogate Pairs...... 192 Figure G-8. Box-cox Transformation of Dependent Variable of THg concentrations ...... 192 Figure G-9. OLS Model Scatterplot for all HUC 8 Subbasins based on the Dataset of 63 Surrogate Pairs...... 193 Figure G-10. 95 Percent Prediction Interval (fixed effect only) for LME Model Regression with Seasons as an Additional Fixed Effect and with Sites as a Random Effect for the 63 Instream THg-TSS surrogate sample pairs. THg concentrations are ng/L and TSS Concentrations are mg/L...... 193 Figure G-11. 95 Percent Prediction Interval (fixed effect only) for LME Model Regression with Sites as a Random Effect for the 63 Instream THg-TSS surrogate sample pairs. THg concentrations are ng/L and TSS Concentrations are mg/L...... 194 Figure G-12. Diagnostic Plot for OLS Model for all 63 Instream THg-TSS Surrogate Samples. The residuals versus fitted values plot checks for the assumptions of linearity...... 195 Figure G-13. Diagnostic Plot for LME Model with Seasons as an Additional Fixed Effect and Sites a Random Effect for all 63 Instream THg-TSS Surrogate Samples...... 196 Figure G-14. Diagnostic Plot for LME Model with Sites as a Random Effect for all 63 Instream THg-TSS Surrogate Samples...... 197 Figure G-15. Empirical Cumulative Distribution of TSS Concentrations across all sites...... 198 Figure G-16. Empirical Cumulative Distribution of TSS Concentrations with Reductions in the Upper Quartile Range (75th percentile) of TSS Concentrations across all sites...... 199 Figure I-1. Sampling locations for more recent (2013-2017) Lower Willamette total mercury data...... 207 Figure I-2. Sampling locations for more recent (2012-2019) Tualatin total mercury data...... 207 Figure I-3. Watersheds simulated in HSPF for the Tualatin HUC08 (17090010)...... 210

State of Oregon Department of Environmental Quality x Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

This page intentionally left blank

State of Oregon Department of Environmental Quality xi Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 Executive Summary

Water quality standards are in place to protect people from high levels of mercury exposure when eating fish. The Willamette River and many of its tributaries do not currently meet water quality standards for mercury and are included on Oregon’s list of impaired waters under Clean Water Act §303(d). Mercury fish consumption advisories are in place throughout the Willamette Basin. The Clean Water Act requires a Total Maximum Daily Load, or TMDL, to be developed for waters that are on the 303(d) list. A TMDL is the amount of a pollutant, in this case mercury, that can be present in a waterway and still meet water quality standards. In Oregon, the Oregon Department of Environmental Quality is responsible for developing and implementing TMDLs.

A Willamette Basin Mercury TMDL was first issued in 2006. This proposed revised Willamette Basin Mercury TMDL identifies sources of mercury and how much mercury needs to be reduced to meet water quality standards. The revised evaluations included in this TMDL are more robust, using linked models and significantly more data than the 2006 TMDL.

By far, the greatest source of mercury in the basin is from atmospheric deposition, which is mercury in the air falling onto the land or into the water. The mercury in air originates mainly from national and global sources rather than from sources in Oregon. Once mercury is deposited on the landscape, the major pathways to streams are erosion of sediment-bound mercury and surface runoff.

Of the many different types of land use that exist within the Willamette Basin, forestry, agriculture and urban uses comprise most of the area within the basin. Management actions on these land uses influence the amount of mercury from these sources that reach streams and rivers in the basin. Point source discharges, such as sewage treatment plants or industries, contribute significantly less mercury to streams than nonpoint sources, such as runoff from logging roads and agricultural fields. The contribution from point sources determined in this draft is less than was estimated in the 2006 TMDL.

This TMDL specifies needed mercury reductions in two ways: nonpoint source load allocations and point source wasteload allocations. The accompanying Water Quality Management Plan describes DEQ’s plan for implementing these allocations and actions to reduce mercury. Mercury minimization measures will be applied for both point and nonpoint source activities, with primary focus on reducing runoff and erosion from nonpoint source activities and urban stormwater. Effectiveness of these measures will be tracked, evaluated and improved, as warranted, to meet the standards.

There is inherent uncertainty in any analysis of complex physical, chemical and biological processes of mercury generation and transport in a very large riverine system like the Willamette Basin. DEQ used models and techniques that incorporate large amounts of water quality and land use data and information specific to the Willamette Basin. However, in order to evaluate these processes on this scale, the models and techniques used simplify these complex processes. To minimize uncertainty and ensure the allocations are protective enough to meet the applicable water quality standards, DEQ used several conservative approaches in multiple components of the analysis. For example, DEQ pooled data on a basin-wide scale, used median concentrations in calculations rather than means, assumed all mercury was available to be methylated and then checked results on a subbasin scale using data collected more recently than that used in the TMDL modeling. Despite these conservative measures, the actual

Oregon Department of Environmental Quality 1 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

response of the system to the management measures is likely to vary from predictions, to some degree. This drives the importance of ongoing monitoring, reporting, and adaptation of targets and measures over time.

DEQ convened a 25-member advisory committee in 2017 to provide input on source identification, allocations and development of an implementable Water Quality Management Plan for achieving required source reductions. Over the course of nine meetings, the committee listened to many hours of technical presentations, participated in complex discussions and provided valuable input from their constituent groups to aid DEQ in reaching complicated technical and policy decisions, which are reflected in this revised draft TMDL.

The final TMDL and Water Quality Management Plan document needed mercury reductions and planned implementation through permits, best management practices, conservation practices and other management strategies to reduce mercury entering streams to the degree that will be necessary for the eventual attainment of the mercury criterion and, ultimately, full restoration of the beneficial use of fish consumption and protection of aquatic life and wildlife throughout the Willamette Basin. The goals, objectives and approaches of this TMDL are consistent with the requirements of the federal Clean Water Act and Oregon water quality laws and implementing regulations.

Oregon Department of Environmental Quality 2 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 1. TMDL Introduction

This Willamette Basin Mercury TMDL was developed in cooperation between DEQ and EPA, along with EPA’s watershed contractor, TetraTech, to address mercury impairments in the Willamette Basin to achieve the Oregon criterion for fish tissue concentrations of methylmercury. Mercury is a pollutant of global concern due to its widespread distribution in the environment and accumulation in aquatic biota. Methylmercury is a potent neurotoxin in humans and other vertebrates.

The draft TMDL document depends directly of the work done in the TMDL Technical Support Document, prepared by EPA’s contractor, Tetra Tech, and presented as Appendix A: Technical Support Document. Sections 1 through 12 of this document summarize the work of the technical support document and describes how this work was used in the decision process of the TMDL. Cross-references are provided to relevant sections in the TMDL Technical Support Document from which reported results were drawn. The WQMP appears in in Section 13 of this document and provides the framework for describing management efforts that will be put into action to attain the proposed Willamette Basin Mercury TMDL. This framework builds upon existing point and nonpoint source implementation plans to outline a management approach for reducing mercury from all land uses in the basin. Reasonable assurance is provided in Section 14, which describes an accountability framework for implementation of sector-specific or source-specific management strategies that will be carried out through regulatory or voluntary actions and monitoring and reporting on progress in meeting TMDL goals. 1.1 TMDL Authority

DEQ is the Oregon state agency responsible for implementing the Clean Water Act in Oregon. The Clean Water Act allows for the delegation of responsibility for many Clean Water Act programs to states, which in Oregon is administered by the Oregon Environmental Quality Commission through Oregon Revised Statute. The Commission granted the DEQ Director authority to develop TMDLs and issue them as orders (Oregon Administrative Rule 340-042- 0060). DEQ was granted authority by the Commission to implement TMDLs through Oregon Administrative Rule 340-042 with special provision for agricultural lands and non-federal forest lands governed by the Agricultural Water Quality Management Act and the Forest Practices Act, respectively. The Clean Water Act requires EPA to approve or disapprove TMDLs that states submit. When a TMDL is officially submitted by a state to EPA, EPA has 30 days to take action on the TMDL. If EPA disapproves a TMDL, the CWA gives EPA 30 days to establish the replacement TMDL. 1.2 General TMDL approach

A TMDL, or total pollutant load to a waterbody, is the sum of individual waste loads allocated to point sources, load allocations assigned to non-point sources and loads assigned to background. The amount of pollutant that a waterbody can receive and still meet the applicable water quality standard is referred to as the “loading” or “assimilative capacity” of the waterbody, and it is calculated as part of the TMDL. Loading from all pollutant sources must not exceed the loading or assimilative capacity (also referred to as the TMDL) of a waterbody and must include an appropriate margin of safety.

Oregon Department of Environmental Quality 3 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Load allocations are portions of the loading capacity that are attributed to either natural background sources, such as soils, or from non-point sources, such as urban, rural agriculture, or forestry activities. Wasteload allocations are portions of the load that are allotted to point sources of pollution, such as sewage treatment plants or industries. The wasteload allocations are used to establish effluent limits in discharge permits. Allocations can also be reserved for future uses, also known as the “reserve capacity.” Allocations are quantified measures that assure water quality standards will be met and may distribute the pollutant loads between nonpoint and point sources. This general TMDL concept is represented by the following equation:

TMDL = Wasteload Allocation + Load Allocation + Reserve Capacity + Margin of Safety

Together, these elements establish the mercury loads necessary to meet the applicable water quality standards for mercury and protect human health, wildlife, aquatic life and other beneficial uses.

The components of the Willamette Basin Mercury TMDL are summarized in Table 1-1.

Table 1-1. Summary of Willamette Basin Mercury TMDL Components TMDL Element Legal Authority Description Name and Location OAR 340-042- This TMDL covers all State of Oregon perennial and 0040(4)(a) intermittent streams in the Willamette Basin (Hydrologic Unit Code (HUC) 170900) (Figure 1-1).

Water Quality OAR 340-042- Toxic Substances: (1) Toxic Substances Narrative. Standards 0040(c) Toxic substances may not be introduced above natural OAR 340-041- background levels in waters of the state in amounts, 0033(1); concentrations, or combinations that may be harmful, OAR 340-041- may chemically change to harmful forms in the 8033(1) Table 30; environment, or may accumulate in sediments or OAR 340-041- bioaccumulate in aquatic life or wildlife to levels that 8033(3) Table 40 adversely affect public health, safety, or welfare or OAR 340-041-0004 aquatic life, wildlife or other designated beneficial uses.

(2) Aquatic Life Numeric Criteria. Levels of toxic substances in waters of the state may not exceed the applicable aquatic life criteria as defined in Table 30 under OAR 340-041-8033. Table 30: Mercury (total) freshwater aquatic life chronic criteria 0.012 ug/L;

(3) Human Health Numeric Criteria. The criteria for waters of the state listed in Table 40 under OAR 340- 041-8033 are established to protect Oregonians from potential adverse health effects associated with long- term exposure to toxic substances associated with consumption of fish, shellfish and water. Table 40: Methylmercury fish tissue criteria 0.040 mg/Kg (See Section 2. Applicable water quality standards) (4) Further degradation will be prevented by following Oregon’s Antidegradation Policy (OAR 340-041-0004) that provides the requirements for making decisions when considering any increases in mercury load to

Oregon Department of Environmental Quality 4 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

TMDL Element Legal Authority Description streams and rivers in the Willamette Basin that DEQ has authority to regulate Designated OAR 340-042- Designated Beneficial Uses, Willamette Basin. The Beneficial Uses 0040(4)(c); TMDL developed to meet the mercury water quality OAR 340-041- standards in OAR 340-041-041-0033(1), OAR 340-041- 0002(17); 8033(1) Table 30, and OAR 340-041-8033(3) Table 40 OAR 340-041-0340 is expected to be protective of all Willamette Basin Table 340A designated beneficial uses including: fish and aquatic life, wildlife and hunting, fishing (See Section 2. Beneficial uses). Pollutant OAR 340-042- Methylmercury and total mercury (See Section 2. 0040(4)(b) Beneficial uses)

TMDL Target The TMDL target is 0.14 ng/L of total mercury in the water column based on the simulated bioaccumulation of methylmercury for Northern Pikeminnow (See Sections 6.1 and 6.1.2) Loading Capacity OAR 340-042- The amount of total mercury in the water column that 0040(4)(d), the Willamette Basin can receive and still meet water 40 CFR 130.2(f) quality standards. Total Mercury loading capacity: 43.36 g/day (See Section 7.1. Loading capacity)

Excess Load OAR 340-042- The difference between the load for total mercury in the 0040(4)(e) Willamette Basin and the loading capacity of the Willamette Basin. Total Mercury excess load: 318 g/day (See Section 7.2. Excess load)

Sources OAR 340-042- Assessment of mercury sources in the Willamette Basin. 0040(4)(f) Point Sources: 4 percent Nonpoint Sources: 96 percent (See Section 9. Source assessment)

Waste Load OAR 340-042- Wasteload allocations for point sources: Allocations 0040(4)(g), • Wastewater Discharge Sector - 10 percent 40 CFR 130.2(h) reduction g Total Mercury/day • Permitted Stormwater Sector - 75 percent reduction g Total Mercury/day (See Section 10.2. Wasteload allocations for point sources) Load Allocations OAR 340-042- Load allocations for nonpoint sources: 0040(4)(h), • General Nonpoint Source Sector: (Forestry, 40 CFR 130.2(g) Agriculture, Water Impoundments, Water Conveyance Entities, Background) – 88 percent reduction g total mercury/day • Legacy Metals Mining Sector – 95 percent reduction g total mercury/day • Non-Permitted Urban Stormwater – 75 percent reduction g total mercury/day • Atmospheric Deposition – 11 percent reduction g total mercury/day (See Section 10.1. Load allocations for nonpoint sources)

Oregon Department of Environmental Quality 5 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

TMDL Element Legal Authority Description Reserve Capacity OAR 340-042- Total Mercury allocations for future growth and new or 0040(4)(k), expanded sources. An explicit reserve capacity of 1 40 CFR 130.2(h) percent g total mercury/day was used (See Section 12. Reserve capacity) Margin of Safety OAR 340-042- An implicit margin of safety that accounts for uncertainty 0040(4)(i), related to the TMDL analyses. 40 CFR 130.7(c)(2) (See Section 11. Margin of safety)

Seasonal Variation OAR 340-042- Seasonal variation and critical conditions for mercury 0040(4)(j), bioaccumulation, loading and sources were considered. 40 CFR 130.7(c)(2) (See Section 8. Seasonal variation and critical condition)

1.3 History of Willamette Basin Mercury TMDL Starting in 1998, DEQ began identifying various waterbodies in the Willamette Basin as impaired due to elevated levels of mercury in fish tissue. This earlier work culminated in the development of the 2006 Total Maximum Daily Load for mercury in the Willamette Basin (Oregon Department of Environmental Quality, 2006). A TMDL is a means for implementing additional controls needed to restore and maintain the quality of water resources (U.S. Environmental Protection Agency, 1991). TMDLs represent the total pollutant loading that a waterbody can receive without exceeding water quality standards. In 2011, Oregon adopted human health criteria based on a revised fish consumption rate of 175 g/day (previously 17.5 g/day), which resulted in a mercury fish tissue criterion of 0.04 mg/kg, as methylmercury (previous TMDL target was 0.3 mg/kg). In 2012, EPA’s approval of DEQ’s Willamette Basin Mercury TMDL was challenged and in 2017, the U.S. District Court for Oregon ordered that the 2006 TMDL remain in place while EPA and Oregon evaluate the TMDL for reissuance consistent with the updated standard, initially by April 2019 and subsequently extended to November 2019.

The 2006 TMDL was developed and presented as an interim solution and at several points suggested the need for additional data collection and refinements of the analytical approach. The April 11, 2017 U.S. District Court, District of Oregon order (Northwest Environmental Advocates v. United States Environmental Protection Agency, 2017) stated that EPA and Oregon must complete a revised TMDL within two years. This order also adopts the earlier findings of Magistrate Judge Acosta (Northwest Environmental Advocates v. United States Environmental Protection Agency, 2016). Judge Acosta’s findings highlighted DEQ and EPA declarations stating that the revision will “require analysis of factors affecting mercury pollution, including potential multiple sources, bioaccumulation patterns, and changes in the types of mercury being released and transformed in the entire complex river system,” and the existing modeling “must be revised and incorporate all the new data related to mercury that has been gathered since the first TMDL” (page 14). Based on DEQ and EPA’s completion of that described analysis, this TMDL targets the currently applicable Oregon fish tissue criterion for protection of human health and is expressed in terms of a daily load. DEQ also considered developing wasteload allocations for individual point sources, but determined that aggregated wasteload allocations were most appropriate. This determination aligns with the Clean Water Act, federal regulations, and EPA guidance specific to TMDL development when air deposition is the dominant source of mercury. Other important technical considerations that contributed to

Oregon Department of Environmental Quality 6 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 this decision were that combined wastewater point source loads contribute only one percent of the total basin load of mercury and the same minimization measures will be applied, regardless of the allocation being individual or aggregated. Distribution of the estimated 4.0 g/day of allowable mercury across 327 permitted wastewater discharges would be inequitable and difficult for implementation when the aggregated approach would be easier to implement and result in the same environmental benefit. The updated TMDL builds upon the existing TMDL modeling analysis, while also making substantial improvements. 1.4 Name and location Per OAR 340-042-0040(a), this element describes the geographic area for which the TMDL is developed. This TMDL covers all perennial and intermittent streams in the Willamette Basin including the mainstem Willamette and Middle Fork Willamette.

Oregon’s Willamette River is approximately 187 miles in length and is the 13th largest river in the lower 48 states in terms of stream flow. The average annual discharge to the Columbia River of 22.73 million acre-feet is nearly 15 percent of the total Columbia River flow (Kammerer, May 1990). The mainstem Willamette River flows to the north and through the City of Portland, Oregon near its confluence with the Columbia River. The Willamette Basin is bounded by the Cascades mountain range to the east and the Coast Range to the west. 1.4.1 HUCs, subbasins, and watersheds The mainstem Willamette River includes flow from 12 major tributaries (Figure 1-1), collectively referred to as the Willamette Basin. DEQ uses the United States Geological Survey system of hydrologic delineation known as Hydrologic Unit Codes for all TMDLs. OAR 340-042-0030(4) defines Hydrologic Unit Code as “a multi-scale numeric code used by the USGS to classify major areas of surface drainage in the United States. The code includes fields for geographic regions, geographic subregions, major river basins and subbasins. The third field of the code generally corresponds to the major river basins named in OAR Chapter 340, Division 41. The fourth field generally corresponds to the subbasins typically addressed in TMDLs.” HUCs often cross political or jurisdictional boundaries such as state, county or city limits and land use boundaries. The twelve tributaries of the mainstem Willamette River roughly correspond to the 12 HUC8 watersheds (Figure 1-1).

Table 1-2. HUC8 codes and corresponding watershed names in the Willamette Basin. HUC8 Code Watershed name 17090001 Middle Fork Willamette River 17090002 Coast Fork Willamette River 17090003 Upper Willamette River 17090004 McKenzie River 17090005 North Santiam River 17090006 South Santiam River 17090007 Middle Willamette River 17090008 Yamhill River 17090009 Molalla-Pudding River 17090010 Tualatin River 17090011 Clackamas River 17090012 Lower Willamette River

Oregon Department of Environmental Quality 7 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Figure 1-1.Willamette Basin

Oregon Department of Environmental Quality 8 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

1.4.2 Land use Land use of the Willamette Basin is predominantly forest with a mixture of agricultural and developed land. The most recent national data sets, derived from remotely sensed data, were used to create land use data for the revised TMDL. The details of the data set development and analysis are in Appendix A: Technical Support Document. The land use categories in the original data sources were aggregated to create more general land use categories that were used in the TMDL. The areas of the general land use categories and the associated percent of the basin’s total area are listed in Table 1-3. The spatial distribution of the general land use categories in the Willamette Basin is shown in Figure 1-2. As can be seen in Figure 1-2, most of the forest is in the mountainous regions along the boundaries of the basin with agriculture, developed and the other land uses nearer the river network and toward the lower elevation regions of the basin. An important consideration is that the land use categories do not directly correspond to the land management. The different reasons for this limitation are discussed in Appendix A: Technical Support Document. For example, the agriculture category may not capture all of the different types of agriculture occurring in the Basin. The agriculture land use category used in the TMDL is the row crops land cover from the national data sets. Other types of agriculture occurring in the Willamette Basin may be grouped under other categories used in the TMDL. For example, pasture and hay areas in the original data source were aggregated under the grassland category used in the TMDL. Another example is orchards that may be grouped under shrubland in the TMDL. Also, the forest land use corresponds generally to the type of land cover and does not directly relate to the management of the area. In other words, forest land use does not directly relate to forestry practices and management that may be occurring in any particular area. This is also true for shrubland, which includes recently harvested forest lands with other land cover. The specific date ranges of the data sources are provided in Appendix A: Technical Support Document. The lack of land management information within the different land use categories is a limitation that affected the source assessment and the subsequent allocations of mercury in the TMDL.

Table 1-3. Land Use Areas and Percentages in Willamette Basin. Land Use Total Area (square miles) Percent of Total Area Agriculture 912 8.0% Barren 102 0.9% Developed 923 8.0% Forest 5,920 51.6% Grassland 1,902 16.6% Shrubland 1,412 12.3% Water 103 0.9% Wetland 192 1.7% Total 11,466 100%

Oregon Department of Environmental Quality 9 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Figure 1-2. Land use distribution in Willamette Basin. Adapted from figure in Appendix: Technical Support Document.

Oregon Department of Environmental Quality 10 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

1.4.3 Climate Season and elevation are the biggest drivers of precipitation in the Willamette Basin. Annual precipitation in the basin is generally greatest between October and March. July and August are typically the driest months with less than 5 percent of total annual precipitation. Elevation plays an important role in the total amount of precipitation because the higher the elevation the greater the total precipitation. For example, the Willamette valley floor may receive 40 to 50 inches annual precipitation compared to the mountainous regions which may receive nearly 200 inches. The Coast Mountains receive more total precipitation, which arrives in the form of rain and some snow pack, while the precipitation in the Cascades comes primarily from snow pack.

1.4.4 Stream Flow The most important characteristic to note about stream flow in the Willamette Basin is that it is highly modified by dam and reservoir operations. Congress passed 15 flood control acts between 1938 and 1974 that affect the Willamette Basin and are implemented by the US Army Corps of Engineers. The purpose of the USACE dams is to provide flood control, navigation, hydroelectric power, and water in summer for irrigation and recreation. Dam operations have dramatically changed the natural flow patterns of the Willamette Basin streams by reducing peak flows in winter and artificially augmenting summer low flows. While these changes are beneficial for flood control, navigation, recreation, and irrigation, there are unintended consequences that influence water quality.

Oregon Department of Environmental Quality 11 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 2. Beneficial uses

Oregon’s beneficial uses are defined in OAR 340-042-0040(4)(c) as those uses of water that the state has identified for waters of the state. The beneficial uses of waters of the state are identified in state statute as the Oregon Environmental Quality Commission adopts by rule beneficial uses by basin. Water quality standards are also adopted by the commission to protect the most sensitive beneficial uses. The beneficial use(s) that is most sensitive to impairment by the pollutant or pollutants addressed in the TMDL will be specified.

This TMDL identifies the beneficial uses in the TMDL geographic area and is developed to protect the most sensitive beneficial uses.

According to OAR 340-041-0340, water quality in the Willamette Basin must be managed to protect a range of beneficial uses (see Table 340A, August 2005). The TMDL was developed to meet applicable water quality standards to protect the most sensitive beneficial uses impaired by mercury, which are: Fish and Aquatic Life; Wildlife and Hunting; and Fishing (fish consumption). The beneficial use of fishing applies to the entire mainstem Willamette River and its tributaries. Meeting water quality standards for the most sensitive beneficial uses will be protective of all other uses.

As well as lack of attainment of water quality criteria, multiple fish consumption advisories issued for the Willamette Basin by the Oregon Health Authority indicate that this beneficial use is not currently being attained. The revised TMDL for mercury is designed to restore the beneficial use of fishing to the Willamette River and its tributaries.

The Oregon Health Authority is the state agency responsible for issuing fish consumption advisories based on contaminant concentrations in fish. Although DEQ reviews fish consumption advisories for mercury as a basis for listing a waterbody on the state's 303(d) list of impaired waterbodies, the overarching goals between these two evaluations differ. DEQ developed the methylmercury fish/shellfish tissue standard of 0.040 mg/kg to protect the beneficial use of "fishing", which allows people to safely consume up to 175 grams of fish per day (or about 23 8-oz fish meals a month) over their lifetime. The TMDL was developed to meet the beneficial use of “fishing”, rather than how much fish are safe to eat based on the amount of mercury already present in fish tissue, which is the basis for fish consumption advisories.

Fish consumption advisories for mercury are currently in place for bass in all Oregon waters; for all resident fish (except stocked, fin-clipped rainbow trout 12-inches or less) in the Dorena and Cottage Grove Reservoirs; and the entire mainstem Willamette River from its mouth on the Columbia River southward to Eugene, including the Coast Fork Willamette up to the Cottage Grove Reservoir (Oregon Health Authority, 2019). The initial fish consumption advisory for the mainstem Willamette River, dated February 13, 1997, advised the public of elevated mercury levels in the edible fish tissue of bass and northern pikeminnow (squawfish) and recommended specific limits for consumers who eat these fish caught anywhere in the mainstem river system (from the mouth of the river upstream to the Cottage Grove Reservoir). The average level of mercury found in bass and northern pikeminnow was 0.63 mg/kg. To determine if there is risk to human health from consuming mercury in fish, Oregon Health Authority first compares mean concentration found in tissue of a particular species to current screening values of 0.2 mg/kg for vulnerable populations and 0.6 mg/kg for the general population (Oregon Health Authority, 2016; Hillwig, 2019). This is similar to the value used (0.35 mg/kg) for the fish consumption advisories in place for the 2006 TMDL.

Oregon Department of Environmental Quality 12 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 3. Applicable water quality standards

Water Quality Standards: This TMDL was developed to meet the applicable water quality standards for protection of the most sensitive beneficial uses. The applicable water quality standards are: OAR 340-042-0040(c); OAR 340-041-0033(1); OAR 340-041-8033(1) Table 30; OAR 340-041-8033(3) Table 40.

The current methylmercury fish tissue criterion in Table 40 is 0.04 milligrams of methylmercury per kilogram of fish tissue. It is based on a fish consumption rate of 175 grams/day and has been in place since 2011. This replaces the previous TMDL target referenced in Chapter 3 of the 2006 Willamette Basin Mercury TMDL. The currently applicable rules are excerpted below: 1. Toxic Substances Narrative. Toxic substances may not be introduced above natural background levels in waters of the state in amounts, concentrations, or combinations that may be harmful, may chemically change to harmful forms in the environment, or may accumulate in sediments or bioaccumulate in aquatic life or wildlife to levels that adversely affect public health, safety, or welfare or aquatic life, wildlife or other designated beneficial uses. 2. Aquatic Life Numeric Criteria. Levels of toxic substances in waters of the state may not exceed the applicable aquatic life criteria as defined in Table 30 under OAR 340-041-8033. Table 30: Mercury (total) freshwater aquatic life chronic criterion 0.012 ug/L; 3. Human Health Numeric Criteria. The criteria for waters of the state listed in Table 40 under OAR 340-041-8033 are established to protect Oregonians from potential adverse health effects associated with long-term exposure to toxic substances associated with consumption of fish, shellfish and water. Table 40: Methylmercury fish tissue criteria 0.040 mg/kg 4. Antidegradation. Further degradation will be prevented by following Oregon’s Antidegradation Policy (OAR 340-041-0004) that provides the requirements for making decisions when considering any increases in mercury load to streams and rivers in the Willamette Basin that DEQ has authority to regulate

The 2006 TMDL development included modeling that generated a bioaccumulation factor for the Willamette Basin for several species of fish. In addition, the TMDL developed a translator to convert the dissolved methylmercury water concentration to a water concentration for total mercury in ng/L. Through these procedures, the TMDL derived water column targets for total mercury in ng/L based on the bioaccumulation factor for the most sensitive species modeled, the Northern pikeminnow (Ptychocheilus oregonensis), at the 50th percentile of the fish tissue data distribution. In 2018, an EPA contractor conducted the modelling needed to update the water concentration value based current methylmercury criterion of 0.04 mg/kg. The revised water column concentration of 0.14 ng/L total mercury was used to update the TMDL.

The Oregon Health Authority cites federal reports by EPA in 2001 and USGS in 2008 in estimating that 90 percent of the mercury found in fish tissue is in the methylated form (Oregon Health Authority, 2016). The average fish tissue concentration for mercury in a number of fish species in the Willamette Basin currently exceeds the 0.040 mg/kg criterion. The current freshwater ‘acute’ criterion for mercury is 2.4 ug/L and the freshwater ‘chronic’ criterion is 0.012 ug/L (12 ng/L) for the protection of aquatic life (as presented in the Table 30 Water Quality Criteria Summary; OAR 340-041-0033).

This TMDL is being developed to meet the human health criterion, which is the most stringent, and will therefore be protective of other uses and applicable criteria. In July 2017, EPA approved California statewide water quality standards for mercury in fish tissue for the

Oregon Department of Environmental Quality 13 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

protection of aquatic life and wildlife dependent species. The approved prey fish objective for protection of wildlife was 0.05 milligram mercury per kilogram fish. Based on this EPA action, DEQ anticipates the Oregon fish tissue criterion of 0.04 milligram methylmercury per kilogram fish would be protective of human health and wildlife uses.

This TMDL also ensures that downstream and adjacent waters of neighboring states are protected. The Columbia River is a shared waterbody with the State of Washington at the mouth of Willamette River and has standards set by both Washington and Oregon for their respective jurisdictions of the river. Development of the TMDL water column concentration target relied on several conservative assumptions, which are detailed in Section 11 on margin of safety. In addition to the margin of safety, DEQ considered the lower flows of the Willamette River entering the Columbia River with dilution factors ranging from: minimum = 2.5; median = 11.2; maximum = 43.6, to conclude that Washington’s slightly more stringent methylmercury fish tissue concentration of 0.03 mg/kg will also be met through implementation of the TMDL.

Oregon Department of Environmental Quality 14 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 4. Mercury water quality impairments

The latest water quality assessments relative to mercury for the Willamette Basin from Oregon’s 2012 Integrated Report (http://www.deq.state.or.us/wq/assessment/rpt2012/results.asp) identified several segments of the Willamette River and its tributaries as not meeting water quality standards. Those waters currently assessed as requiring a TMDL (Category 5) are shown in Figure 4-1. Waters for which mercury impairments were addressed by the 2006 TMDL are indicated in the last column of Table 4-1. Segments that were listed in Category 5 in the Willamette Basin after the completion of the 2006 TMDL are also addressed in this revised TMDL.

Table 4-1. 303(d) Listings for Mercury in the Willamette Basin 2006 Name Miles HUC 8 Assessed Affected Use Category TMDL Amazon Diversion Canal 0 to 3.9 17090003 2010 Fishing 5 X (A3 Drain) Amazon Creek Diversion 0 to 6.6 17090003 2010 Fishing 5 X Canal Yamhill River 0 to 11.2 17090008 2012 Human health 5 Resident fish and aquatic life; Coast Fork Willamette/ 28.5 to 17090002 2012 Anadromous fish 5 X Cottage Grove Reservoir 31.3 passage; Drinking water Drinking water; Resident fish and 7.3 to Row River/ Dorena Lake 17090002 2012 aquatic life; 5 X 11.9 Anadromous fish passage Clackamas River 0 to 83.2 17090011 2012 Human health 5 Tualatin River 0 to 80.7 17090010 2012 Human health 5 Multnomah Channel 0 to 21.7 17090012 2012 Human health 5 Middle Fork Willamette 0 to 82.2 17090001 2012 Human health 5 River Coast Fork Willamette 0 to 38.8 17090002 2012 Human health 5 X River Coast Fork Willamette 31.3 to Aquatic life; Human 17090002 2012 5 X River 38.8 health McKenzie River 0 to 84.8 17090004 2012 Human health 5 Dennis Creek Aquatic life; Human 0 to 1.4 17090002 2012 5 X health Santiam River 0 to 26.2 17090005 2012 Human health 5 Willamette River 17090003 0 to 186.6 17090007 2012 Human health 5 X 17090012

Note: Information from 2012 Integrated Report as of May 2019. Category 5 = Water quality limited, TMDL needed.

Oregon Department of Environmental Quality 15 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Note: The most recent Integrated Report in Oregon was approved by EPA December 20, 2018, with no changes to mercury impairments in the Willamette Basin. Figure 4-1. Mercury Impairments in the Willamette Basin

Oregon Department of Environmental Quality 16 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 5. Summary of Mercury TMDL development and approach

5.1 Mercury cycling in the environment This TMDL was developed to achieve the Oregon human health criterion for fish tissue concentrations of methylmercury and thereby achieve the other applicable criteria. Mercury in higher trophic level fish is present largely as methylmercury, which is a potent neurotoxin in humans and other vertebrates. Mercury is a pollutant of global concern due to its widespread distribution in the environment and accumulation in aquatic biota. Most releases of mercury into the environment are to the atmosphere in an inorganic form; however, almost all human exposure to mercury is to an organic form, methylmercury, through the consumption of contaminated fish (Eagles-Smith, et al., 2018; Munthe, et al., 2007). Mercury released into the atmosphere has a long atmospheric lifetime (~6-12 months) which allows for its widespread distribution prior to deposition (Lindberg, et al., 2007; Schroeder & Munthe, 1998). As a result, elevated levels of methylmercury in fish tissue occur even in remote ecosystem (Chetelat, et al., 2015; Fitzgerald, Engstrom, Mason, & Nater, 1998; Trip & Allan, 2000). Most of the mercury in fish originates form dietary exposure, with minimal direct uptake by fish from the water (Hall, Bodaly, Fudge, Rudd, & Rosenberg, 1997). Therefore, differences in trophic position, foraging behavior, and diet can have a large impact on how much mercury is present in a given fish species (Driscoll, et al., 2007; Eagles-Smith, et al., 2016a).

Mercury is deposited from the atmosphere via wet (rain and snow) and dry deposition processes (Lindberg, et al., 2007; Wright, Zhang, & Marsik, 2016). Watershed characteristics and associated land use activities have a large influence on how much atmospherically deposited mercury is sequestered in the soil, re-emitted to the atmosphere, or mobilized in surface water (Domagalski, et al., 2016; Eckley, et al., 2016; Hsu-Kim, et al., 2018). In addition to atmospheric deposition, some watersheds may also have point sources of mercury pollution (e.g. mines or other industries that utilize mercury containing materials) and/or may contain soils geologically enriched with mercury that can contribute to mercury levels in water and fish (Domagalski, et al., 2016; Kocman, et al., 2017; Smith, Cannon, Woodruff, Solano, & Ellefsen, 2014). The Willamette Basin receives most of its mercury from atmospheric deposition that originates from trans-Pacific sources (Strode, et al., 2008).

The form of mercury that bioaccumulates in fish tissue is almost exclusively methylmercury (Driscoll, et al., 2007; Munthe, et al., 2007). Therefore, it is important to understand the environmental processes impacting mercury methylation. Anaerobic bacteria perform mercury methylation, which can involve sulfate and iron reducing bacteria as well as methanogenic bacteria (Christensen, et al., 2016; Ullrich, Tanton, & Abdrashitova, 2001; Warner, Roden, & Bonzongo, 2003). The amount of methylmercury produced in the environment is a result of two main factors: 1) the availability of inorganic mercury and 2) the presence and activity of microorganisms capable of methylizing mercury (Marvin-DePasquale, et al., 2009). Typically, only a relatively small amount of the total inorganic mercury present is in a form that is available to methylating microorganisms (Hsu-Kim, et al., 2018; Marvin-DePasquale, et al., 2009). In general, mercury that is tightly bound to sediment or particles is less available for microbial methylation than mercury that is in the dissolved phase and/or bound to dissolved organic carbon (Benoit, Gilmour, Heyes, Mason, & Miller, 2002; Graham, Aiken, & Gilmour, 2012; Hsu- Kim, Kucharzyk, Zhang, & Deshusses, 2013). Factors that affect methylating microorganisms include the presence of anoxic (oxygen-free) conditions, nutrients, terminal electron acceptors

Oregon Department of Environmental Quality 17 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

(such as sulfate, ferric iron, etc.), and organic matter (Hsu-Kim, et al., 2018). Because the microorganisms responsible for methylation require anoxic conditions, methylation tends to occur in sediments and the stagnant waters in wetlands and thermally stratified lakes and reservoirs.

Landscape variables can have a large influence on the fate of atmospherically deposited mercury. For example, when mercury is deposited to a natural forested watershed, the vast majority of the mercury is retained in vegetation and soil and only a relatively small amount is exported in runoff (Balogh, Nollet, & Offerman, 2005; Domagalski, et al., 2016; Hsu-Kim, et al., 2018; Shanley, et al., 2008). The retention of mercury in watersheds is due to its binding to sulfur function groups associated with soil organic matter (Obrist, et al., 2016; Skyllberg, Xia, Bloom, Nater, & Bleam, 2000). Watershed land uses such as forestry, agriculture, and urbanization have all been shown to decrease the retention of atmospherically deposited mercury and enhance its mobilization into adjoining waterbodies (Eckley, et al., 2018; Eckley & Branfireun, 2009; Hsu-Kim, et al., 2018). Water impoundments, such as reservoirs have been shown to have elevated fish methylmercury levels relative to natural lakes and free-flowing rivers (Willacker, et al., 2016). The variables that have been shown to enhance fish methylmercury levels in reservoirs include a shift from a lotic to lentic foodweb, the development of anoxic conditions in the bottom water, the accumulation of organic matter and inorganic mercury from settling particles, and water-level fluctuations enhancing sulfate and organic carbon cycling (Eckley, Luxton, Goetz, & McKernan, 2017; Hsu-Kim, et al., 2018). Overall, these landscape variables can have a large impact on how much atmospherically deposited ends up bioaccumulating in fish tissue (Eagles-Smith, et al., 2016b). 5.2 Mercury TMDL approach Determining the TMDL linkage between total mercury loads and attaining fish tissue concentrations of methylmercury to protect human health is complicated because of the many intervening kinetic and transport processes. Methylmercury is produced under anoxic conditions, which can occur within a river or watershed. Within a river, methylmercury production mostly occurs within the sediment, with the quiescent water of backwater channels potentially having higher rates of methylation. Within a watershed, wetlands or areas with saturated soils can often provide important locations for methylmercury production. The relative importance of internally produced (within the waterbodies and their sediments) or externally produced (within soils and groundwater prior to reaching waterbodies) sources of methylmercury has not been assessed for the Willamette Basin. Methylmercury monitoring data are available primarily from the water column. The simplified conceptual framework used in this TMDL is that the long-term average methylmercury concentration in the water column depends on total mercury concentrations in the sediment, which in turn, depend on rates of total mercury loading from upstream. The complex transformations between different forms of mercury are not explicitly simulated; rather, they are approximated by an empirical relationship between observed methylmercury and total mercury in the water column, as described in the following sections.

The TMDL report depends directly of the work done in the TMDL Technical Support Document, prepared by EPA’s contractor, Tetra Tech, and presented as Appendix A: Technical Support Document. This TMDL document summarizes the work of the technical support document and how this work was used in the decision process of the TMDL. Cross-references are provided in this TMDL document to sections in the technical support document from which reported results were drawn.

Oregon Department of Environmental Quality 18 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Linked models were used to determine the relationship between fish tissue concentrations of methylmercury (or MeHg) and various sources of total mercury (or THg) in the Willamette Basin. These models estimated the load capacity, calculated the excess load, identified sources and linked the sources of total mercury in the Willamette Basin to the concentrations in the river system. The first model used was the Food Web Model. The Food Web Model simulates bioaccumulation and predicts fish tissue concentrations of methylmercury based on exposure concentrations in the water column. This relationship is then used to estimate mercury exposure concentrations that are consistent with attaining the water quality standard for methylmercury in fish tissue. Next, a mercury translator equation was used to estimate the total mercury in the water river system. At this point, the load capacity of total mercury for a target fish species was estimated. The difference between the load capacity and the median of observed total mercury concentrations from throughout the Willamette Basin was used to calculate the excess load. The third model used was a Mass Balance Model, which evaluated contributions of various sources of total mercury within the Willamette Basin and linked the sources to observed total mercury in water of the river system. The models linked together and the different forms of mercury among those links is shown below in Figure 5-1. The three model components were calibrated for the current conditions using recent data for methylmercury in fish tissue, forms of mercury in water and various input data that characterizes the sources of total mercury in the Willamette Basin and the processes that control the fate and transport of the sources. Details on each modeling step are presented below.

MeHg in MeHg in THg in THg Fish Water Water Sources Tissue

Food Mercury Mass Web Transator Balance Model Model

Legend Forms of Mercury (Hg) MeHg – methylmercury THg – total mercury

Model component

Figure 5-1. Linked models and forms of mercury used to develop the Willamette Basin Mercury TMDL.

Oregon Department of Environmental Quality 19 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 6. Explanation of models and current mercury load

The conceptual framework used in this TMDL is that the long-term average methylmercury concentration in the water column depends on rates of total mercury loading from upstream. Linked models populated with basin-scale data were used to connect fish tissue concentrations of methylmercury to sources of total mercury in the Willamette Basin. These models estimated the load capacity, calculated the excess load, identified sources and linked the sources of total mercury to the concentrations in the river system. These evaluations were all conducted using data pooled at the basin-scale to ensure water quality and fish tissue data were matched, to minimize potential variability due to known data differentials at the subbasin (HUC8) scale and as a better indicator of the central tendancy of the data, which represents the appropriate exposure scenarios in meeting the human health methylmercury criterion. Details of each component of the models and calculation of the current mercury loads are discussed in the remainder of this section.

6.1 Explanation of models DEQ’s approach to determining the load capacity for the revised TMDL consists of two fundamental methodological components: a basin-specific aquatic food web model to estimate methylmercury biomagnification; and a translator equation that calculates the total mercury concentration in the water column, using methylmercury biomagnification information. This approach allows relation of the fish tissue concentration of methylmercury to a load capacity based on total mercury concentration in the water column of the Willamette River and its tributaries. The total mercury concentration can then be related to total mercury sources entering the Willamette River and its tributaries. The discussion below contains a more detailed technical presentation of the analytical tools utilized in the development of the load capacity for this mercury TMDL.

MeHg in MeHg in THg in Fish Water Water Tissue

Food Web Mercury Model Translator

Figure 6-1. Conceptual model for estimating load capacity from fish-tissue MeHg to water column THg.

Oregon Department of Environmental Quality 20 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

6.1.1 Food Web Model A model of the magnification of mercury through the food web was used to relate fish tissue methylmercury concentrations to the total mercury concentrations in the surface waters of the Willamette Basin. This was done using the Food Web Model, which was the same model used in the 2006 TMDL. The Food Web Model is a peer reviewed model (Hope, 2003) that simulates the concentration of mercury in different fish species located at different levels in the food chain (or trophic levels). The basic concept in the Food Web Model is that as one fish species eats other fish species with low mercury concentration in their tissue, mercury will concentrate in the tissue of the predator at levels greater than the prey. The species and the respective trophic levels used in this analysis are listed in Table 6-1. These are the fish species for which total mercury target concentrations in the water were estimated consistent with attaining the water quality standard for levels of methylmercury in the fish tissue. These species reside in Willamette Basin waters year round, so are better indicators of mercury accumulation based on pollutant sources in Oregon. The Food Web Model was calibrated using water quality and fish tissue data collected in the Willamette Basin since 2002. This dataset was documented in the Willamette River Basin Mercury Data Summary (Schmidt, 2018). The methods and results of the model update and calibration are presented in the TMDL Technical Support Document in Appendix A: Technical Support Document.

Table 6-1. Fish Species and Associated Trophic Levels used in the Food Web Model Species Trophic Level Bluegill (BLU) Level 3 Carp (CAR) Level 3 Cutthroat Trout (CTT) Level 3 Rainbow Trout (RBT) Level 3 Largescale Sucker (LSS) Level 3 Largemouth Bass (LMB) Level 4 Smallmouth Bass (SMB) Level 4 Northern Pikeminnow (NPM) Level 4

The Food Web Model is a probabilistic model and allows for the incorporation of uncertainty and risk into the analysis. The Food Web Model is run many times (10,000 times) and input data for each run is taken from random selections of values from the statistical distributions of the input data for the Willamette Basin. Estimation of these statistical distributions was a main component of the calibration of the Food Web Model. The specific information, statistical distributions and estimated parameters for the updated Food Web Model are in Section 3 of Appendix A: Technical Support Document. The output data from the Food Web Model was used with the translator equation to obtain datasets of total mercury concentrations in the waters of the Willamette Basin for each of the fish species listed in Table 6-1.

The Food Web Model also estimated biomagnification for each fish species, which is a measure of how much methylmercury will concentrate in the tissue of fish species within the food web. Several statistics of the distributions of the biomagnification values from the Food Web Model are listed in Table 6-2. These estimates were compared to national values developed by EPA for trophic levels 3 and 4 (U.S. Environmental Protection Agency, 2001) and are shown in Figure 6-2. The estimated values fall near or within 95% confidence intervals by trophic level on the national values (see Figure 6-2). DEQ considered this sufficient indication that the biomagnification estimates specifically for the Willamette Basin were within the acceptable ranges of values for similar food webs in the US.

Oregon Department of Environmental Quality 21 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Table 6-2. Biomagnification (L/kg) estimates from Food Web Model for fish species. Fish Species Median 5th Percentile 95th Percentile Bluegill 1.90E+06 1.34E+06 7.76E+06 Carp 2.52E+06 1.92E+06 6.77E+06 Cutthroat Trout 4.72E+05 3.00E+05 2.23E+06 Rainbow Trout 2.27E+06 1.40E+06 1.15E+07 Largescale Sucker 2.16E+06 1.63E+06 5.97E+06 Largemouth Bass 5.02E+06 3.48E+06 1.83E+07 Smallmouth Bass 7.66E+05 4.62E+05 4.30E+06 Northern Pikeminnow 1.81E+06 1.18E+06 7.21E+06

Figure 6-2. Comparison of Food Web Model biomagnification estimates to national values bioaccumulation factors values. Source: (U.S. Environmental Protection Agency, 2010a). Blue band represents the 5th to 95th percentile of the national BAF estimates. Relationship between BMF and BAF are discussed more in TSD. 6.1.2 Mercury translator equation DEQ used the information from the Food Web Model to relate total mercury concentrations in water to dissolved methylmercury in fish tissue for higher trophic level fish species. Based on paired methylmercury and total mercury samples in the Willamette Basin, about five percent of the total mercury concentration observed in water is methylmercury. Therefore, a majority of total mercury is inorganic. However, under certain environmental conditions, methylation can

Oregon Department of Environmental Quality 22 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

occur converting stored inorganic mercury to methylmercury. DEQ established a total mercury concentration target for the TMDL to reduce methylation potential and because loads from sources in the basin are predominately in the inorganic mercury form.

In the same manner in the 2006 TMDL, DEQ used an empirical translator equation to relate dissolved methylmercury and total mercury in the water column using more recent observed data. The translator was applied to estimate basin-wide target total mercury concentrations based on critical dissolved methylmercury exposure concentrations for different fish species. DEQ used this empirical relationship to represent the complex, non-linear methylation process that depends on temperature, carbon, sulfur, reduction/oxidation conditions and other variables. The details of the translator equation development are given in Section 4 of Appendix A: Technical Support Document.

As explained in the TMDL Technical Support Document, consideration was needed around issues of data censoring of total mercury and dissolved methylmercury data, data not collected at the same time and place, as well as spatial and seasonal variability of the translator equation. These issues were addressed as follows. To address data-censoring concerns, all of the data was used in the translator development and methods that provide robust statistical approaches were used for the censored portion of the data. The issue of the data not collected at the same time and place was addressed using the medians of the dissolved methylmercury and total mercury data, which were calculated for the entire period the data was collected. To address spatial variation, medians of dissolved methylmercury and total mercury concentrations were calculated for each HUC8 in which data was collected. Because the ratios of these medians tended to be consistent among the HUC8s, DEQ considered this sufficient evidence that a single translator equation would be effective. For seasonal variation, translator equations for two seasons (June through October and November through May) were considered. Different slopes for the translator equations were found for the two seasons with a larger slope for the warmer months. The slope estimates and associated statistics are listed in Table 6-3. The increased uncertainty of the slope estimates for the seasons, as opposed to the entire year, is evident in the larger slope standard error and wider confidence intervals compared to the single slope estimate for the entire year (see Table 6-3). The final translator equation used the entire year slope estimate (see Figure 6-3).

Table 6-3. Slope estimates and associated statistics for translator equation. Slope Lower 95% Upper 95% Slope Slope p- Season Estimate1 Confidence Confidence Standard value (unitless) Limit Limit Error Entire Year 0.0160 0.0147 0.0174 0.0006 <0.0001 June through 0.0347 0.0300 0.0393 0.0021 <0.0001 October November 0.0070 0.0057 0.0083 0.0057 <0.0001 through May 1 Weighted least-squares method was used to estimate the slope. The weights were the number of observations for each HUC8 used to calculate the medians of dMeHg and THg.

Oregon Department of Environmental Quality 23 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Figure 6-3. Final Mercury Translator Model: Aggregated, Year-round, Zero-Intercept Model by HUC8 Weighted by Sample Size.

MDL is minimum detection limit of methylmercury laboratory method. The triangles are the median pairs of dissolved methylmercury and total mercury for each HUC8.

The final step is to estimate the total mercury target level for each fish species based on the updated fish tissue concentration. The details of the estimation of the total mercury target levels for the fish species considered in Section 4.3 of Appendix A: Technical Support Document. The same formula used by Hope in the 2006 TMDL was used for this calculation and is given as: = · , · 𝑇𝑇𝑇𝑇 where: 𝑇𝑇𝑇𝑇𝑛𝑛 � � 𝐶𝐶𝐶𝐶 𝑀𝑀𝑀𝑀 𝑛𝑛th TLn is the total mercury target level for𝐵𝐵𝐵𝐵𝐵𝐵 the n fishΩ species (ng/L), TC is the revised fish tissue criterion for MeHg in fish (0.040 mg/kg), th BMFME, n is the biomagnification factor for the n fish species (L/kg – see Table 6-2), Ω represents the Mercury Translator (0.0016 from slope estimate in Table 6-3), and CF is a conversion factor (1 · 106 ng/mg).

All of the values for the parameters in the equation for TLn are the same for all the fish species except the biomagnification factor, which was simulated using the Food Web Model for each fish species. Furthermore, 10,000 simulations were run using randomly selected values from multiple input parameter distributions to get a biomagnification factor distribution for each species, and ultimately a species-specific distribution of target total mercury concentration. The medians of the target total mercury levels for each fish species were calculated from these data sets. The confidence intervals of the median target total mercury levels were estimated directly

Oregon Department of Environmental Quality 24 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

using a bootstrapping method, which does not rely on assumptions about the statistical distributions of the datasets.

The final estimates of the target total mercury levels and associated confidence intervals for each fish species are listed in Table 6-4. The 0.14 ng/L total mercury target level in water needed to achieve the fish tissue criteria in Northern Pikeminnow was used to calculate a single load capacity of total mercury for the entire Willamette Basin. DEQ selected this target level for the development of the TMDL because this was the lowest total mercury concentration among the fish species and considered the most protective. This is the same fish species that was used to develop the target in the 2006 TMDL. This median concentration is 40 percent lower than the mean concentration for Northern Pikeminnow of 0.23 ng/L. Using the median is consistent with DEQ’s development of the methylmercury fish tissue criterion and adds an additional measure of conservation intended to address regional differences in the data set and ensure that water quality standrads will be met in all listed reaches within each subbasin.

Table 6-4. Species-specific Surface Water THg Target Levels to Meet a Fish Tissue Concentration of 0.040 mg/kg MeHg. Median THg Lower 95% Confidence Upper 95% Confidence Species Target Levels Limit on Median Limit on Median (ng/L) Bluegill 0.32 0.27 0.38 Carp 0.37 0.33 0.41 Cutthroat Trout 1.11 0.98 1.31 Largemouth Bass 0.22 0.19 0.26 Largescale Sucker 0.42 0.38 0.47 Northern Pikeminnow 0.14 0.12 0.15 Rainbow Trout 0.58 0.50 0.69 Smallmouth Bass 0.35 0.31 0.40

6.1.3 Linkage analysis using Mass Balance Model As detailed in the TMDL Technical Support Document, the linkage of mercury sources to the levels of total mercury in the Willamette River and its tributaries was analyzed using a Mass Balance Model. Figure 6-1 shows this connection in a conceptual model form. The 2006 TMDL mass balance used simplified delivery ratios for total mercury loads delivered from atmospheric deposition and soil erosion to the waters of Willamette Basin. This revision of the mercury TMDL used a detailed source characterization and a comprehensive watershed model called Hydrological Simulation Program – FORTRAN (HSPF) to more robustly simulate the generation and transport of total mercury within the Willamette Basin. The main vectors for total mercury that DEQ considered were dissolved forms from precipitation, permitted discharges, surface runoff and groundwater, along with mercury attached to eroded soils and stream sediment. The Mass Balance Model estimated the fluxes of these vectors into and through the Willamette Basin. The processes simulated in the HSPF model allowed for estimates of the distribution of mercury in the water column and sediment.

Oregon Department of Environmental Quality 25 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Total Mercury in Total Mercury Water Sources

Mass Balance Model

Figure 6-4. Conceptual model of linking total mercury source in the watershed to total mercury in the streams and rivers of the Willamette Basin.

DEQ used the results of the Mass Balance Model to estimate the overall mercury load in the Willamette Basin and distribute the load to different source categories. These source categories are at the top of Figure 6-5 and included permitted point sources and nonpoint sources. The nonpoint sources were broken down further by source, land use, and methods of transport. These methods of transport were direct to water, overland flow, resurfacing groundwater, and erosion of soil. An example of how a nonpoint source was represented is mercury from wet atmospheric deposition in rain onto agricultural land (a pervious land use type) moved to stream network by overland flow. The distribution of the mercury sources allowed DEQ to investigate the relative contributions and to develop allocations. The relative contributions of the sources were calculated as the ratio of the source load to the total load estimated using the Mass Balance Model. The development of the model and how the Willamette Basin was represented is described in detail in Appendix A: Technical Support Document.

Oregon Department of Environmental Quality 26 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Figure 6-5. Conceptual Framework for the Total Mercury Mass Balance Model. Figure taken from Appendix: Technical Support Document.

The Willamette Basin was broken up into sub-basins in the HSPF model. This allowed for the source of mercury and the transport methods to be separated spatially in more detail than the HUC8s. The sub-basins are shown in Figure 6-6 and within each sub-basin, sources were represented using model HSPF model elements referred to as hydrologic response units. These units combine topography, soils, land use, weather and other relevant information to create homogeneous units that are simulated and combined in the HSPF model to represent movement of water and sediment within and through the Willamette Basin.

Oregon Department of Environmental Quality 27 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Figure 6-6. Existing HSPF Model Domain for the Willamette Basin. Taken from Appendix: Technical Support Document.

Oregon Department of Environmental Quality 28 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

An example of the process to generate the hydrologic response units is shown in Figure 6-7. Upland processes and pathways of transport (e.g., surface runoff, groundwater resurfacing) simulated by the HSPF model were used to characterize flow and sediment export from nonpoint sources in the watershed, and paired with mercury source information to estimate nonpoint source mercury loads. Next the relationships of the different mercury sources to the pathways described in Figure 6-5 were applied using output from the HSPF model simulations and mercury source information. The development of the mercury source information for the Mass Balance Model required a comprehensive source characterization, which is summarized in the following section.

Figure 6-7. Schematic of Model Hydrologic Response Unit (HRU) Development. Taken Appendix: Technical Support Document.

Oregon Department of Environmental Quality 29 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

6.1.4 Nonpoint source input data development The principal sources included in the nonpoint source analysis were atmospheric deposition, soil and groundwater. As shown in Figure 6-5 above, these sources follow various pathways to get into Willamette Basin streams. Little information was available for concentrations of mercury specific to the groundwater aquifers of the Willamette Basin. Several large scale studies were available to estimate mercury loads in the Willamette Basin from the atmosphere and in the soils. These loads are mixtures of background and anthropogenic sources. 6.1.4.1 Atmospheric deposition Both wet and dry deposition rates of mercury were estimated using previous studies for input to the Mass Balance Model. Wet deposition of mercury occurs in a dissolved form in precipitation. Dry deposition is associated with dust and other small particulates. No significant spatial variations were found across the Willamette Basin for either form of deposition. Single values were estimated for the wet and dry deposition and are listed in Table 6-5. The wet deposition load was derived from wet deposition concentration data and surface runoff volume. The dry deposition rate was used to calculate the deposition of mercury on days without precipitation and a build-up/wash-off approach was used to estimate delivery to the stream network.

Table 6-5. Estimated Wet and Dry Atmospheric Deposition Rates in the Willamette Basin. Deposition Type Value

Wet deposition of total mercury concentration in precipitation 6.05 ng/L Dry deposition of total mercury 4.24 µg/m2/year

Research indicates that most of the atmospheric sources of mercury deposited in the Willamette Basin originate outside the basin. Air quality modeling of sources within Oregon was not undertaken in the TMDL to quantify atmospheric deposition of mercury from such sources due to their relatively small contribution to the total of atmospheric deposition. See section 5.3.1 of the TMDL Technical Support Document for full discussion of estimating atmospheric deposition of mercury in the Willamette Basin.

6.1.4.2 Soil Mercury attached to soil can be delivered to Willamette Basin streams by a number of processes. Previous studies were used to estimate the total mercury levels in the soils throughout the basin. The method used to account for the mercury level from soils was to estimate potency factors (mass of mercury per mass of sediment expressed as ug/kg) for use in the Mass Balance Model.. A description of potency factors can be found in Section 5.3.2 of the TMDL Technical Suport Document. The potency factors were multiplied by the sediment loads estimated in the Mass Balance Model to get the mercury loads from soil erosion and sediment delivered to streams. These loads can then be associated with locations in the Basin at the HUC8 level and land uses. The potency factors varied with geology, soil properties and land use type. The main effect of land use was the retention and re-emission rates related to the vegetative cover (e.g. forest, shrub, or cultivated land). Using this approach, several potency factors for soils were estimated and are listed in Table 6-6. The forest and shrub land cover potency factor varied across the HUC8s in the basin. Sufficient amounts of data were not available to determine whether there was significant variation between HUCs for land cover other than forest or shrub and single values were used for each land use for the entire basin. The potency factors were multiplied by the sediment load for the land cover transported to

Oregon Department of Environmental Quality 30 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 streams. The sediment loads were taken from the watershed model of the Willamette Basin that is part of the Mass Balance Model. Although many of the sources of mercury in the soil may be from geology or other sources not directly related to human activities, the erosion and transport of soil to Willamette Basin streams are, in part, controllable. Therefore, these pathways are addressed in the TMDL and WQMP.

Table 6-6. Soil total mercury potency factor estimates for Willamette Basin Land Cover HUC8 THg Potency (µg/kg) Forest and Shrub 17090001 49.7 Forest and Shrub 17090002 48.2 Forest and Shrub 17090003 85.4 Forest and Shrub 17090004 60.7 Forest and Shrub 17090005 80 Forest and Shrub 17090006 79.7 Forest and Shrub 17090007 96.8 Forest and Shrub 17090008 105.1 Forest and Shrub 17090009 90.2 Forest and Shrub 17090010 115.9 Forest and Shrub 17090011 77.3 Forest and Shrub 17090012 111 Cultivated Land All 36.7 Herbaceous Upland All 23.3 Other All 30.1

6.1.4.3 Groundwater Mercury can occur in dissolved form in groundwater. The mercury load dissolved in groundwater was estimated using simulated groundwater flow from the watershed model and estimated total mercury concentration in groundwater. Limited data are available on groundwater mercury concentrations. No studies were found that accurately characterize mercury in groundwater in the Willamette Basin, or elsewhere in the Pacific Northwest. There were some samples collected from groundwater sources near Superfund cleanup sites of the mining areas in the Coast Fork Willamette River watershed, which were low (near 1 ng/L) or below the detection limit of 0.5 ng/L. Some additional data for forested watersheds in the Coast Range of Oregon also indicated that total mercury concentrations were low in groundwater (less than 1 ng/L). A value of 1 ng/L was selected as the concentration of total mercury in groundwater for the entire Willamette Basin for the TMDL. While 1 ng/L is a low concentration, groundwater entering streams is a large component of the total flow of water in the basin. As such, this resulted in large loads of total mercury (approximately 17 percent of the total source load to the stream network) estimated from groundwater contributions. Improved evaluation of the concentration of total mercury in groundwater in the basin will be included in the monitoring stragetgy for the TMDL to reduce the uncertainty associated with this source. 6.1.5 Estimation of instream delivery of mercury and reservoir processes Not all of the mercury loads that enter Willamette Basin streams are delivered to the outlet of the basin, as it empties into the Columbia River. Some of the load is lost due to atmospheric and in stream processes throughout the stream network, including reservoir processes related to several impoundments located in the basin (as described in Section 9.2.2 below). The details of the methods and the results are discussed in Appendix A: Technical Support Document.

Oregon Department of Environmental Quality 31 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Losses of mercury during transit in the stream network were represented with an exponential decay model calibrated to observed data. Losses of mercury specifically due to reservoirs were estimated using observed mercury concentrations in both inflow and outflow of reservoirs, along with flow volumes from some of the reservoirs. These estimated losses for the reservoirs were then applied to the simulated mercury loads from the HSPF model. Despite acknowledgement of these losses, DEQ opted to use the at-source rather than the delivered loads, as a measure of conservatism in calculating the existing loads, loading capacity, excess loads, needed reductions and allocations. As noted in comparison of Figures 5-17 and 5-18 in the TMDL Technical Support Document, this resulted in 65 percent larger existing loads. Thus, larger reductions are needed to meet the in-stream target. In part, this conservatism addresses uncertainty and ensures that, with implementation of the TMDL, water quality criteria will be met in listed reaches in all subbasins. 6.2 Current total mercury load estimation for the Willamette Basin The Mass Balance Model was used to estimate the total load of mercury and the contributions of those loads by source categories. The way the Mass Balance Model was constructed for this TMDL allows for investigating loads from the HUC8 geographic areas and even within the HUC8s. However, the loads for the entire Willamette Basin were evaluated, as another conservative measure to ensure that water quality criteria will be met in all subbasins through implementation of the TMDL. The rational for focusing on the entire basin is discussed in more detail in Section 7.2 and explains the excess load and load capacity calculation. The estimated loads and the relative contribution of each source category are listed in Table 6-7. The majority of the load (greater than 95 percent) is from nonpoint sources, and the permitted point sources account for less than five percent. The estimated total load of 361.3 g/day is the value used for the current load of total mercury and is used in the calculations of the excess load and load capacity.

Table 6-7. Estimate total mercury loads for source categories from Mass Balance Model for Willamette Basin. Estimated Load of Relative Contribution to Source Category Total Mercury (g/day) Total Load

Nonpoint Sources

Surface Runoff of atmospherically 118.0 32.7% deposited mercury

Resurfacing Groundwater 60.6 16.8%

Direct deposition to open water 5.9 1.6%

Erosion of mercury containing soils 154.6 42.8%

Urban DMAs (without MS4 permits) 2.5 0.7%

Legacy mine discharges 4.0 1.1%

Oregon Department of Environmental Quality 32 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Point Sources

Sewage Treatment Plant discharges 3.2 0.9%

Industrial discharges 1.2 0.3%

Permitted Stormwater discharges 11.3 3.1%

Total Load 361.3

Note: These are loads at the point of entry into the stream network and not loads delivered to the mouth of the Willamette.

Oregon Department of Environmental Quality 33 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

7. Loading capacity and excess load

7.1 Loading capacity Summarizing OAR 340-042-0040(4)(d) and 40 Code of Federal Regulations 130.2(f), loading capacity is the amount of a pollutant or pollutants that a waterbody can receive and still meet water quality standards. Modeled estimation of the amount of total mercury in the water column that the Willamette Basin streams can receive and still meet water quality standards was developed. The TMDL Technical Support Document indicates that the basin-wide loading capacity for Willamette Basin is:

• Total Mercury loading capacity: 43 g/day

7.2 Excess load In accordance with OAR 340-042-0040(4)(e), the excess load calculation evaluates, to the extent existing data allow, the difference between the actual pollutant load in a waterbody and the loading capacity of that waterbody. The basin-wide excess load for the Willamette Basin is:

• Total mercury excess load: 318 g/day

The excess load was calculated by first calculating the required reduction. The required reduction was calculated using the target total mercury concentration and observed total mercury concentrations from throughout the Willamette Basin. Excess load for the Willamette Basin, then, is the difference between the basin-wide existing loads of total mercury and the basin-wide loading capacities of total mercury. Required reduction can be inferred from existing total mercury water column concentrations:

= 1 × 100 𝑇𝑇𝑇𝑇 𝑅𝑅 − where is the required percent reduction, is 𝐸𝐸the𝐸𝐸 target surface water total mercury concentration (ng/L), and is the existing surface water total mercury concentration calculated from observed𝑅𝑅 data collected during the period𝑇𝑇𝑇𝑇 of 2002-2014. Boxplots of the observed total mercury data are shown in𝐸𝐸 𝐸𝐸Figure 7-1.

Oregon Department of Environmental Quality 34 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Figure 7-1. Boxplots of observed total mercury concentrations by HUC8 (See Table 1-2) and medians of entire data sets with data from Coast Fork and without. The data for the Coast Fork HUC8 (17090002) is highlighed with the HUC8 name above the boxplot. No observed data were available for South Santiam (17090006).

The existing basin-wide surface water total mercury concentration ( ) was calculated as the median of the observed total mercury concentrations. DEQ considered calculating the required reduction for each HUC8 and then calculating load capacity by HUC8.𝐸𝐸𝐸𝐸 However, the number of observed data was not evenly distributed among the HUC8, and there were no data available for the South Santiam (17090006). The different number of observations for the different HUC8s can be seen in Figure 7-1, for example the Coast Fork (17090002) and the Lower Willamette (17090012) have many more observations compared to the McKenzie (17090004). The differences in the number of observations between HUC8s could result in additional uncertainty in the estimated medians of total mercury for the HUC8s that have less data available compared to the HUC8s that have more data available. To minimize this uncertainty, DEQ pooled all of the HUC8 data together and calculated a single median for the existing surface water total mercury concentration for the entire Willamette Basin.

Next, DEQ considered the conditions related to the mining districts in the Coast Fork HUC8 (17090002) to be a poor representation of the remaining Willamette Basin. The effects of legacy mercury contamination is apparent in the observed data for the Coast Fork, as highlighted in Figure 7-1. DEQ compared the median for the entire dataset to the median of the data excluding the total mercury concentration data collected in the Coast Fork HUC8. The differences in HUC8 medians are shown in Figure 7-1. The median for the data set excluding the Coast Fork

Oregon Department of Environmental Quality 35 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

data is 1.2 ng/L and the median for the data set including the Coast Fork data is 1.6 ng/L. The estimated median total mercury concentration increased by 25 percent when the Coast Fork data were included.

DEQ used the median of the surface water total mercury data, excluding the Coast Fork HUC8 data, to better represent the existing conditions for the entire Willamette Basin and used the median value of 1.2 ng/L to calculate the required reduction.

As shown in Figure 7.2, the medians in the Coast Fork, Tualatin and Lower Willamette subbasins are above the basin-wide median. However, each is within the distribution of the majority of the data used to calculate the basin-wide median and therefore, need not be treated differently from the rest of the basin. To confirm that achieving the calculated reductions would achieve the fish tissue methylmercury criterion in all basin waterways, DEQ undertook additional evaluations in the three subbasins with median concentrations above the basin-wide median.

Figure 7-2. Comparison of the individual medians of the HUC08 total mercury data to the distribution of the data represented by the boxplot of the data used to calculate the basin-wide existing condition of 1.2 ng/L.

To get a better estimate of the total mercury existing condition for these HUC8s, DEQ evaluated more recent data than was used in the TMDL modeling analysis. DEQ compiled total mercury data collected in the mainstem of the Tualatin River between 2012 to 2019 and in the mainstem of the Lower Willamette River reach between 2013 to 2017. This data is presented in Appendix I. As detailed in Section 14.1.4, significant mercury reductions actions have occurred in these target basins since collection of the data used in the TMDL analyses (through 2011). As such, this recent data is more representative of current conditions at the outlet of these HUC8s with implementation of the 2006 TMDL. As shown in Figure 7-3, the medians of the more recent data from these subbasins more closely align to the basin-wide median. This limited analysis indicates that mercury reduction measures are effectively achieving in-stream responses in these subbasins and adds confidence that the basin-wide TMDL approach is protective of all basin waters and its implementation will achieve the methylmercury criterion in all listed reaches in all subbasins.

Oregon Department of Environmental Quality 36 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Figure 7-3. Comparison of the individual total mercury medians of the Lower Willamette and Tualatin HUC08s using total mercury data collected in the mainstems between 2012 and 2019 to the distribution of the data represented by the boxplot of the data used to calculate the basin-wide existing condition of 1.2 ng/L.

The required basin-wide reductions that are needed in order to meet the target total mercury concentrations were calculated for each of the fish species listed in Table 7-1. The required reduction of 88 percent for the Northern Pikeminnow was the largest (see Table 7-1) and this was selected as the required reduction to be used to calculate the excess load. Rationale for selecting the required reduction for the Northern Pikeminnow is discussed further in Section 11. Margin of safety.

As detailed in Section 9.2.3, legacy mines are responsible for significant contributions of mercury to Coast Fork subbasin streams, which, in part, results in this higher median concetration. As explained in Section 10 on Allocations, DEQ assigned a high reduction allocation for legacy mine-related mercury sources. Together with the nonpoint source and point source reductions, the effective reduction requirement in the Coast Fork is 93%, which is higher than the basin-wide reduction. Implementation of these reductions is currently occurring under EPA’s Superfund authority and Abandoned Mine Lands programs led by Bureaue of Land Management, U.S. Forest Service and DEQ, as discussed in Sections 13.1. Therefore, DEQ is confident that implementation of this higher effective reduction in the Coast Fork subbasin will ensure the methylmercury fish tissue criterion will be met in all listed reaches there.

Oregon Department of Environmental Quality 37 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Table 7-1. Required reductions of the existing median surface water total mercury concentration of 1.2 ng/L to meet fish species specific surface water total mercury target levels. Surface Water Total Mercury Target Levels Required Fish Species (ng/L) Reduction to Meet Fish Tissue Concentration Northern Pikeminnow 0.14 88% Largemouth Bass 0.22 82% Bluegill 0.32 73% Smallmouth Bass 0.35 71% Carp 0.37 69% Largescale Sucker 0.42 65% Rainbow Trout 0.58 52% Cutthroat Trout 1.11 7%

The basin-wide excess load was calculated next by using the required reduction and the estimated daily load for the Willamette Basin, which was calculated using the Mass Balance Model. The required percent reduction was applied directly to the current load to estimate the excess load: = ×

where is is the excess load in g/day𝐿𝐿𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸, is𝑅𝑅 the 𝐿𝐿required𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 reduction in load (expressed as a fraction), and is the current load in g/day estimated using the Mass Balance Model, 𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸 which was𝐿𝐿 estimated to be 361.3 g/day. The𝑅𝑅 total mercury excess load for the Willamette Basin is: 𝐿𝐿𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶

= 0.88 × 361.3 = 317.94 𝑔𝑔 𝑔𝑔 𝐿𝐿𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸 The basin-wide load capacity can then be calculated𝑑𝑑𝑑𝑑𝑑𝑑 as the remain𝑑𝑑𝑑𝑑𝑑𝑑ing load: = = 361.3 317.94 = 43.36 𝑔𝑔 𝑔𝑔 𝑔𝑔 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿 𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 𝐿𝐿𝐶𝐶𝐶𝐶𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟 − 𝐿𝐿𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸 − = 43.36 𝑑𝑑𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑𝑑𝑑 𝑔𝑔 𝑜𝑜𝑜𝑜 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿 𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 This basin-wide load capacity of 43.36 g/day corresponds 𝑑𝑑𝑑𝑑𝑑𝑑to the 88 percent required reduction and was used to calculate allocations and remaining parts of the TMDL equation.

Oregon Department of Environmental Quality 38 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 8. Seasonal variation and critical condition

Per OAR 340-042-0040(4)(j) and 40 Code of Federal Regulation130.7(c)(2), TMDLs must also identify seasonal variation and the critical condition. Seasonal variation, of the processes and sources effecting methylmercury concentrations in fish tissue, was considered by using observed data collected throughout the year and through the use of models that incorporate seasonality. The Food Web Model did not explicitly address seasonality, but the data used to calibrate the model were from different times during the year. Observed total mercury and methylmercury data collected throughout the year were also used to develop the mercury translator equation, which adequately addresses any seasonal variation of conditions that drive the conversion of total mercury to methymercury and total mercury loading. A single mercury translator equation for the entire year represents these varied conditions.

The Mass Balance Model was used to explicitly incorporate the seasonal variation of the sources and processes that control the methylmercury concentration in fish tissue. DEQ determined that the Mass Balance model simulation ran at an hourly time-step adequately represents the seasonal variation of sources and processes in the Willamette Basin, including flows regulated by reservoir management.

Bioaccumulation of mercury in fish is a long-term process related to the intake of prey containing methylmercury and, to a lesser extent, methylmercury concentrations in the water. It takes several years for fish to accumulate enough methylmercury to exceed the fish-tissue criterion, resulting in the issuance of fish consumption advisories. Critical conditions, such as periods of critical low flow, are typically used for permitting pollutant discharges to achieve less acute or shorter-term impacts. Because the bioaccumulation of methylmercury is a long-term process, there were no specific conditions that could be considered critical in this TMDL.

Oregon Department of Environmental Quality 39 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 9. Source assessment

As noted in OAR 340-042-0040(4)(f) and OAR 340-042-030(12), a source is any process, practice, activity or resulting condition that causes or may cause pollution or the introduction of pollutants to a waterbody. This section identifies the mercury sources and estimates, to the extent existing data allow, the amount of actual mercury loading from existing sources. Sources of mercury to streams include point and nonpoint sources. Specific sources are described below and are subsequently assigned allocations. By mass, nonpoint sources are the major sources of mercury in the Willamette Basin. “Nonpoint sources are diffuse or unconfined sources of pollution where wastes can either enter, or be conveyed by the movement of water, into waters of the state” OAR 340-41-0002 (42). 9.1 Air emissions 9.1.1 Oregon permitted mercury air emissions

Within the Willamette Basin there are currently 221 sources that are permitted, either by federal Title V Clean Air Act permits or Oregon Air Contaminant Discharge permits, of which 145 have reported mercury emissions. In 2016, approximately 31.8 kg/yr of measurable mercury was emitted from these permitted facilities, which are included in Appendix F. Only three of these facilities emitted more than 1 kg/yr and a single facility emitted more than 70 percent of the total permitted mercury emissions within the Willamette Basin in 2016. In the remainder of the state, there are approximately 101 permitted sources of mercury emitting approximately 37.3 kg/yr.

DEQ tracks these permitted air emissions and is requiring controls for significant emissions under recently adopted Cleaner Air Oregon rules (OAR 340-245). However, air deposition modeling was not undertaken for this TMDL, so loads deposited due to these air emissions have not been quantified and are not assigned a specific allocation in the TMDL.

9.1.2 National and global mercury emissions trends An existing study for global and North American mercury transport (Seigneur, 2004) was used in evaluating potential mercury air deposition sources. According to this study, 37 percent of mercury sources from human activities deposited in the Willamette Basin from the atmosphere were from outside of North America. In 2019, the United Nations Environment Programme released the “Global Mercury Assessment 2018,” which updated assessments produced at roughly five year intervals since 1990, the last being in 2013. The updated assessment found that mercury emissions continue to decline by 1 to 2 percent per year in North America and Europe, but are increasing in Asia, where approximately 49 percent of global anthropogenic mercury air emissions originate (United Nations Environment Programme, 2019). The United Nations assessment asserts that legacy mining mercury sources from activities up until the end of the 19th century continue to contribute more mercury to soils and waters than all 20th century industrial sources combined. The UN assessment also notes that the potential for remobilization of mercury from these legacy mining sources complicates analysis of fate and transport. In addition, the assessment confirmed a long lag time for reductions in mercury emissions to show responses because methylation of legacy mercury deposited on soils and into aquatic systems will continue (United Nations Environment Programme, 2019).

Oregon Department of Environmental Quality 40 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Significant reductions of atmospheric sources of mercury may not be possible by DEQ because most sources originate outside of Oregon and DEQ’s regulatory control. However, atmospheric deposition is the primary source of mercury in surface runoff and limiting the amount of mercury from the atmosphere traveling across the land in surface runoff can be addressed in the Willamette Basin by DEQ. 9.2 Nonpoint sources Nonpoint sources are diffuse or unconfined sources of pollution where wastes can either enter, or be conveyed by the movement of water, into waters of the state (OAR 340-41-0002 (42)). For purposes of this TMDL, nonpoint sources include activities associated with forestry, agriculture, water impoundments, water conveyances, non-permitted urban and rural stormwater, groundwater and atmospheric deposition to land and water.

Both nonpoint and point source loads are greatly influenced by atmospheric deposition of mercury onto the Oregon landscape. As noted in Figures 5-17 and 5-18 of the TMDL Technical Support Document, modeling indicates that the source categories of surface runoff and sediment erosion together contribute approximately 76 percent of the total mercury load to basin streams. These two source categories are implicated in nonpoint source load contributions due to land use management activities (agriculture, forestry, impoundments, water conveyances, background and non-MS4-permitted urban areas), as well as stormwater point source contributions.

9.2.1 Agriculture, forestry, non-MS4 permitted urban areas, and water conveyances Even though the nonpoint source categories were broken down by land use characteristics in the Mass Balance Model, loads for specific human activities and management could not be separated from natural or background processes and sources. Some examples include the difference between the forest land use category used in the TMDL analysis and forestry management. The forest land use in the TMDL covers natural undisturbed forests and some forestry management. Furthermore, harvests due to forestry are included in the shrub land use in the TMDL, but cannot be separated from shrub land that may be natural, or from disturbed forest that is not the result of harvest, such as fire. In the current TMDL analysis, there are similar limitations in separating Agricultural, Non-Permitted Urban Stormwater areas, and Water Conveyance Systems’ loads from natural or background sources. For these reasons, a single value is used for the load allocation of all the nonpoint sources. 9.2.2 Impoundments Placement of dams in streams creates impoundments or reservoirs. Stream flow in the Willamette Basin is highly modified by dam and reservoir operations. DEQ searched Oregon Water Resources Department records to determine that approximately 414 impoundments of various size and function exist within the Willamette Basin. Many of these impoundments are small and some are ponds, rather than stream impoundments. As explained in the TMDL Technical Support Document, stream impoundments slow water flows, can encourage deposition of sediment and can produce low oxygen conditions that encourage bacterial transformation of mercury to methylmercury. While these processes are complex and difficult to predict, when mercury is present, impoundments can play an important role in mercury cycling (TetraTech, 2019a).

Oregon Department of Environmental Quality 41 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

In assessing impoundments in the Willamette Basin as sources of mercury, an important consideration is whether methylmercuryis transported from reservoirs downstream. As a key component for factoring these effects into its modeling, DEQ, EPA and it’s contractor focused on large reservoirs where information was available. The TMDL Technical Support Document describes considerations around the large impoundments in the basin and how their inclusion in the modeling was determined. Modeled impoundments within the basin include: 11 of the 13 Willamette Basin reservoirs operated by the US Army Corps of Engineers (Figure 9-1) and the North Fork Reservoir, which is the largest sediment trap of the reservoirs operated by PGE in the Willamette Basin (Figure 9-2). The final set of reservoirs represented in the model is shown in Table 9-1.

Oregon Department of Environmental Quality 42 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Figure 9-1. U.S. Army Corps of Engineers Dams and Reservoirs in the Willamette Basin.

Oregon Department of Environmental Quality 43 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Figure 9-2. PGE’s North Fork Dam and Reservoir on the Clackamas River.

Table 9-1. Reservoirs Represented in the Willamette Basin HSPF Model. Reservoir Name River HUC8 Blue River Blue River-McKenzie 17090004 Cottage Grove Coast Fork-Willamette 17090002 Cougar South Fork-McKenzie 17090004 Detroit North Santiam River 17090005 Dorena Row River 17090002 Fall Creek Fall Creek 17090001 Fern Ridge Long Tom River 17090003 Foster South Santiam River 17090006 Green Peter Middle Santiam River 17090006 Hills Creek Middle Fork-Willamette 17090001 Lookout Point Middle Fork-Willamette 17090001 North Fork Clackamas 17090011

9.2.3 Legacy metals mining sources Within the Willamette Basin, there are five abandoned mercury mines, seven mercury prospects (where no extraction or production has taken place) and five districts focused on gold mining (where mercury amalgamation may have taken place). Abandoned mine lands in the basin were identified during mine land investigations conducted by DEQ, US Forest Service and Bureau of Land Management in 2000 to 2010. Table 9-2 lists the 12 mining districts and abandoned mine lands that are currently being assessed and remediated in DEQ’s Cleanup Program, in collaboration with federal partners (EPA, BLM, USFS) with jurisdiction. Thirty-eight individual sites were originally presented in the TMDL Technical Support Document (Table 5-7), which are

Oregon Department of Environmental Quality 44 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

all captured in Table 9-2. Mercury mines or other mines with significant assessment and cleanup are considered as individual mercury sources, while collections of smaller and less studied sites are grouped into mining districts by subbasins.

Table 9-2. Abandoned Mine Lands and Mining Districts Within the Willamette Basin. Primary Mine Name Owner Mine District Drainage County Commodity Black Butte/ Garroutte Creek/ Black Butte Private mercury Elkhead Coast Fork Lane Mercury Mine Mercury Willamette Row River/ Coast Champion Mine Public gold Bohemia Gold Lane Fork Willamette Bohemia Mines Row River/ Coast Public gold Bohemia Gold Lane (10 unpatented) Fork Willamette Bohemia Mines Row River/ Coast Private gold Bohemia Gold Lane (10 patented) Fork Willamette Blue River/ Blue River Mines Public gold Blue River Gold Lane McKenzie Lucky Boy (14 Blue River/ Private gold Blue River Gold Lane patented) McKenzie Quartzville Mines Private gold Quartzville Gold South Santiam Linn (>17 patented) North Santiam Ruth/Amalgamated Private copper, zinc North Santiam Marion Mines North Santiam North Santiam Public gold North Santiam Marion Mines Mines Breitenbush North Santiam Private mercury North Santiam Marion Mercury Mines North Fork North Fork Claims Public mercury Clackamas Clackamas Claims Ames-Bancroft North Fork Private mercury Clackamas Clackamas Mine Claims

Of most relevance is the abandoned Black Butte Mine that is immediately upstream of Cottage Grove Reservoir, the Bohemia gold mining district that is tributary to Dorena Reservoir (where six abandoned sites were identified for further assessment) and the North Santiam River which is a tributary to Detroit Reservoir (where two abandoned sites were identified for further assessment).

As noted in the TMDL Technical Support Document, the Black Butte Mine is the most significant mercury source in the Willamette Basin associated with mining. Furnace Creek, which was significantly impacted by Black Butte Mine activities, was determined to be contributing a substantial percentage of the mercury load to the Coast Fork of the Willamette River. Remediation of Furnace Creek took place in 2018, as part of Superfund cleanup. Evaluation of mercury concentrations and outflows from the Cottage Grove Reservoir was possible, because observed water column data exists for the reservoir area. An example of this relationship is shown in Figure 9-3. The mercury load leaving Cottage Grove Reservoir was estimated by the modeling to be approximately 2.45 kg/yr (TetraTech, 2019a).

Oregon Department of Environmental Quality 45 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Load (kg/month) Elevation (ft)

0.7 800

0.6 790

0.5 780

0.4 770

0.3 760

0.2 750 LAKE LAKE ELEVATION (FT) THG LOAD (KG/MONTH) 0.1 740

0 730 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec MONTH

Figure 9-3. Monthly THg loads from Cottage Grove Reservoir.

The second most significant source of mercury from mining in the basin is associated with the Bohemia Mining District situated among the tributaries of the Row River upstream of Dorena Reservoir. Mercury data is limited in this area, but water samples indicate mercury concentrations flowing into the reservoir are elevated compared to elsewhere in the basin (TetraTech, 2019a). Sediment samples from multiple tributaries indicate mercury contamination from mining sources is the primary cause of elevated mercury in fish tissue in Dorena Reservoir (Hygelund, Ambers, & Ambers, 2001). Mercury concentrations in macroinvertebrates were also more than double those found in the Middle Fork Willamette, which is a nearby subbasin without known mining (Henny, Kaiser, Packard, Grove, & Taft, 2005). The mercury load leaving Dorena Reservoir was estimated by the modeling to be approximately 1.15 kg/yr (TetraTech, 2019a).

Currently, the available data on other abandoned mine lands in the basin is not sufficient to indicate whether these lower priority sites are sources of mercury. DEQ and EPA will continue to assess and remediate, as warranted, the remaining abandoned mine lands within the basin. 9.3 Background and unquantified anthropogenic sources “Background sources include pollutants not originating from human activities and anthropogenic sources of a pollutant that the Department or another Oregon state agency does not have authority to regulate, such as pollutants emanating from another state, tribal lands or sources otherwise beyond the jurisdiction of the state” (OAR 340-042-0030(1)). Much of the atmospheric deposition falls within this definition of background sources. The same is true for groundwater sources where the mercury concentration is a result of geologic properties. In addition, mercury from natural and anthropogenic sources attaches to eroded soils that are delivered as sediment to streams and rivers. However, not all of the surface runoff and eroded soils are background sources and human activities elevate the rates of runoff and erosion processes. The increases in surface runoff and soil erosion resulting from human activities will be addressed in the WQMP and the mercury from these sources will be the focus of reduction efforts.

Oregon Department of Environmental Quality 46 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Atmospheric deposition of mercury is the dominant source of mercury in the Willamette Basin. Much of the mercury coming from the atmosphere is from sources outside of Oregon, which cannot be directly identified or quantified at this time. Sources of mercury from atmospheric deposition that are delivered to streams via the surface runoff pathway also remain unquantified. Mercury from surface runoff and eroded soils coming from anthropogenic activities on forest and shrub lands cannot currently be distinguished from loading from non-disturbed forest lands in this analysis. These limitations were considered when source reductions were being assigned during the load allocation of this TMDL. 9.4 Point sources Point Source means a discernible, confined, and discrete conveyance including, but not limited to, a pipe, ditch, channel, tunnel, conduit, well, discrete fissure, container, rolling stock, concentrated animal feeding operation, vessel or other floating craft, or leachate collection system from which pollutants are or may be discharged but does not include agricultural storm water discharges and return flows from irrigated agriculture (OAR 340-041-0002(46)). DEQ issues NPDES permits for sources that discharge to surface waters according to OAR 340-045- 0015. NPDES permits fall into two categories: general and individual. Existing permit information was obtained from DEQ’s databases for facilities permitted to discharge to Willamette Basin streams.

Only very limited mercury data were available from NPDES-permitted wastewater discharges for development of the 2006 TMDL, so all NPDES-permitted wastewater discharges at the time were considered as a single point source sector with an estimated mercury load of 3.9 percent (or 5 kg/yr) of the total load to the basin. A list of individual facilities was not compiled in the 2006 TMDL and a phased approach was intended to allow collection of additional data. The 2006 modeling estimated a total mercury load of about 2.7 percent (or 3.5 kg/yr) from municipal sewage treatment discharges, using the average flows of 17 facilities ranging from 1.7 to 70 million gallons per day (MGD). The load from industrial discharges was estimated in 2006 at approximately 1.2 percent (or 1.5 kg/yr) from average flows of eight pulp and paper dischargers with flows ranging from 1.7 to 17 MGD. In the current analysis, the wastewater and stormwater point sources together make up just 4 percent of the total mercury load. 9.4.1 Wastewater permits and mercury loads DEQ is delegated by EPA authority under Section 402 the Clean Water Act to issue NPDES permits for waste water discharges to waters of the state. In the Willamette Basin, DEQ issues individual and general permits for wastewater discharged from major and minor municipal sewage treatment plants, major and minor industrial operations that fall under certain Standard Industrial Classification codes, and a variety of general wastewater permit categories. Information on these permit types can be found at: https://www.oregon.gov/deq/wq/wqpermits/Pages/All-Permits-Applications.aspx.

Information on permitted discharges can be found by searching facility name, waterway or permit type in DEQ’s Wastewater Permits Documents Database: https://www.oregon.gov/deq/wq/wqpermits/Pages/Wastewater-Permits-Database.aspx.

9.4.1.1 Municipal sewage treatment plant permits The 2019 modeling, as described in the TMDL Technical Support Document, used mercury data and flows from 23 major sewage treatment facilities that discharge to Willamette Basin streams. Major sewage treatment facilities are defined as facilities with discharges greater than 1 million

Oregon Department of Environmental Quality 47 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

gallons per day, populations greater than 10,000 or with pretreatment programs (see Table 9-3). These data were used to to estimate a municipal discharge mercury load of approximately 0.9 percent (or 1.17 kg/yr or 3.2 g/day) of the total load in the basin. There are also 49 minor sewage treatment facilities that discharge to Willamette Basin streams. Minor sewage treatment facilities are facilities that generally discharge less than 1 million gallons per day. Together, these minor facilities may contribute up to 0.07 percent to the total mercury load in the basin. Cumulatively, flow volumes from minor municipal discharges are comparable to about 9 percent of the cumulative flows from the major facilities or about the same flow volume as one major facility. Due to homogeneity of municipal effluent of any size, potential loads of mercury from all 49 minor facilities are anticipated to be similar to that from a single major facility. Therefore, DEQ determined that the cumulative discharges from minor municipal facilities represent an insignificant portion of the overall contribution of mercury to the watershed.

Table 9-3. Summary of Major NPDES-Permitted Municipal Sewage Treatment Facilities in the Willamette Basin. EPA DEQ Facility Name Receiving Stream Name Permit Permit Number Number Clean Water Services - Rock Creek Tualatin River OR0029777 101144 Metropolitan Wastewater Management Willamette River OR0031224 102486 Commission - Eugene/Springfield City of Salem - Willow Lake Willamette River OR0026409 101145 Clean Water Services - Durham Tualatin River OR0028118 101141 Tri-City Service District - Oregon City Willamette River OR0031259 101168 Clackamas County Service District #1 - Willamette River OR0026221 100983 Kellogg Creek Clean Water Services - Hillsboro Tualatin River OR0023345 101143 Clean Water Services - Forest Grove Tualatin River OR0020168 101142 City of Portland - Tryon Creek Willamette River OR0026891 101614 City of McMinnville Water Reclamation South Yamhill River OR0034002 101062 Facility Albany-Millersburg Water Reclamation Willamette River OR0028801 102024 Facility City of Corvallis Willamette River OR0026361 101714 City of Canby Willamette River OR0020214 101063 City of Wilsonville Willamette River OR0022764 101888 Oak Lodge Water Services District Willamette River OR0026140 100986 City of Woodburn Pudding River OR0020001 101558 City of Dallas Rickreall Creek OR0020737 101518 City of Silverton Silver Creek OR0020656 101720 City of Lebanon South Santiam River OR0020818 101771 City of Newberg - Wynooski Road Willamette River OR0032352 100988 City of Cottage Grove Coast Fork Willamette River OR0020559 101300 City of Stayton North Santiam River OR0020427 101601 City of Sweet Home South Santiam River OR0020346 101657

Oregon Department of Environmental Quality 48 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

9.4.1.2 Industrial wastewater permits For industrial discharges, Standard Industrial Classification code categories of permitted industrial activities were evaluated to identify the potential for mercury to be present in process wastewater discharges. As explained in the TMDL Technical Support Document, the following categories were determined to have the potential for mercury to be present in discharge based on the materials and processes used: timber products; paper products; chemical products; glass, clay, cement, concrete, gypsum products; primary metal industries; fabricated metal products; and electronic instruments. The average mercury concentration derived from facility data from Oregon and other states was applied for these types of facilities in the 2019 modeling. There are 13 active minor industrial permits with identified activities in these categories, which were excluded from the sector because they do not discharge process wastewater. These facilities have other, non-process water discharges such as stormwater, groundwater, non- contact cooling water. As described in Section 9.4.2.2 below, stormwater loads from these facilities are implicit in the municipal stormwater loads. In addition, the modeling used mercury and/or flow data from seven major industrial facilities (defined using EPA’s Industrial Classification worksheet), including several pulp and paper facilities that were operating between 2006 and 2019, but have now ceased operations, and 29 minor industrial facilities discharging to Willamette Basin streams. These data were used to estimate a sector total mercury load of approximately 0.45 kg/yr or 1.23 g/day (which makes up 0.3 percent of the total load to the basin). This estimated load represents contributions from: • Eight major industrial facilities: five pulp and paper (three currently operating); two smelting; one electronics; and • 56 minor industrial facilities: 15 timber; one smelting; four food and beverage; one cooling water; six dairy/hatchery/concentrated animal feeding operation; and 30 not otherwise classified Table 9-4 lists active individual permits for seven major industrial wastewater dischargers and 15 minor facilities which fall into Standard Industrial Classification code categories with activities that have the potential to increase mercury in process wastewater discharge. The SIC Major Group column in the table indicates the main industrial category the facility reported and subcategory information can be obtained in the SIC Manual on the US Department of Labor Occupational Safety and Health website.

Table 9-4. Summary of Individual NPDES-Permitted Industrial Facilities in the Willamette Basin with Activities that could Increase Mercury in their Discharge. SIC EPA DEQ Facility EPA Receiving Facility Name Major Permit Permit Activity Class Water Group Number Number Cascade Pacific Bleached Kraft Willamette 26 Major OR0001074 101114 Pulp, Llc Pulp Mill River Secondary Georgia-Pacific - Willamette 26 Fiber Pulp & Major OR0033405 101488 Halsey Mill River Paper Mill International Unbleached McKenzie Paper Co - 26 Kraft Pulp & Major OR0000515 101081 River Springfield Mill Paper Mill Oregon Titanium Metallurgical, Llc - 33 Manufacturing Major Oak Creek OR0001716 102223 Ati Albany & Forming

Oregon Department of Environmental Quality 49 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

SIC EPA DEQ Facility EPA Receiving Facility Name Major Permit Permit Activity Class Water Group Number Number Tdy Industries, Llc Zirconium - Teledyne Wah 33 Major Truax Creek OR0001112 100522 Production Chang West Linn Paper Paper Willamette Company (Not 26 Major OR0000787 100976 Manufacturing River Operating) Westrock, Fiber Deink Willamette Newberg Mill (Not 26 Pulp & Paper Major OR0000558 101299 River Operating) Mill Ash Grove Cement - Willamette 32 Lime Minor OR0001601 102465 Rivergate Lime River Plant Blue Heron (Not Paper Willamette 26 Major OR0000566 102229 Operating) Manufacturing River South Blast Furnaces Cascade Steel 33 Minor Yamhill OR0027260 101487 & Steel Mills River Evraz Oregon Blast Furnaces Willamette 33 Minor OR0000451 101007 Steel & Steel Mills River North Frank Lumber Co. Sawmills And 24 Minor Santiam OR0000124 101583 Inc. Planing Mills River Abrasive Coffee Lake Fujimi Corporation 32 Minor OR0040339 103033 Products Creek Georgia-Pacific Plastics Murder Millersburg Resin 28 Materials, Minor OR0032107 102603 Creek Plant Synthetics Hollingsworth & Other Wood Willamette Vose Fiber 24 Minor OR0000299 101331 Products River Company Hull-Oakes Sawmills And 24 Minor Oliver Creek OR0038032 101466 Lumber Co. Planing Mills Kingsford Manufacturing Other Wood Patterson 24 Minor OR0031330 102153 Company - Products Slough Springfield Plant Softwood Murphy Veneer, 24 Veneer And Minor Wiley Creek OR0021741 101777 Foster Division Plywood Sanders Wood Products - Rsg Sawmills And Molalla 24 Minor OR0021300 100929 Forest Products - Planing Mills River Liberal Seneca Sawmill Sawmills And 24 Minor Unknown OR0022985 101893 Company Planing Mills Stimson Lumber Sawmills And Scoggins Company - Forest 24 Minor OR0001295 101480 Planing Mills Creek Grove Printed Circuit Sunstone Circuits 36 Minor Milk Creek OR0031127 101015 Boards

Oregon Department of Environmental Quality 50 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

SIC EPA DEQ Facility EPA Receiving Facility Name Major Permit Permit Activity Class Water Group Number Number Weyerhaeuser Softwood Coast Fork Cottage Grove 24 Veneer And Minor Willamette OR0000698 101449 Lumber Plywood River

There are also currently 158 registrants under NPDES general wastewater permits in the basin (36 cooling water, 24 filter backwash, four fish hatcheries, four boiler blowdown, nine petroleum hydrocarbon cleanup, 21 wash water and 60 pesticide application). These activities do not involve mercury and were not identified in the TMDL Technical Support Document’s as activities anticipated to increase mercury in their discharge. The lack of potential for mercury to be present at measurable concentrations coupled with their estimated cumulative flows being very minor, makes the discharges insignificant as an overall contribution to the mercury load within the basin.

As of 2019, there are 33 registrants of the NPDES 700PM for suction dredge mining operating at 46 identified locations within the Willamette Basin. These locations are all clustered in two historical mining areas: Bohemia Mining District above Dorena Reservoir and the Bureau of Land Management’s Quartzville Recreational Mining area. There are 21 registrants with 28 mining site locations on the Row River system, which is a tributary to Dorena Reservoir. There are eight locations on Brice Creek, two on Champion Creek, and 18 on Sharps Creek on the Row River system. There are 12 registrants with 18 mining site locations on Quartzville Creek, which is a tributary to the Middle Fork of the Santiam River. As with other general permit categories, the volume of flows in discharges from these mining activities are anticipated to be insignificant. However, in areas with sediment data confirming mercury contamination, disturbance by suction dredge has a high potential to mobilize and convert mercury to various forms ,which can be transported downstream and methylated (Fleck, et al., 2010; Gray, Hines, Krabbenhoft, & Thoms, 2012; Humphreys, 2005; Marvin-DePasquale, et al., 2009; Marvin- DiPasquale, et al., 2011). This action adds an unquantified but direct source contribution to the loads, which are a function of concentration and flows, and collected behind reservoirs. 9.4.2 Stormwater permits and mercury loads Precipitation falling on urbanized areas within the Willamette Basin can interact with atmospherically-deposited mercury and the resulting stormwater runoff is conveyed to waterways. Stormwater carries polluted runoff from streets, rooftops, parking lots, industrial facilities and construction sites into waterbodies. This runoff can contain pollutants, such as metals, pesticides, PCBs, and PAHs that can harm fish and other aquatic life. Large metropolitan areas, rapidly developing urban areas and high-traffic roadways can be significant sources of these pollutants.

In the 2006 TMDL, urban stormwater loads were not broken out separately from atmospheric deposition and erosion loads, but “runoff of atmospherically-deposited mercury from urban environments was estimated at 5 percent of the total load” (Oregon Department of Environmental Quality, 2006). The lack of site-specific data and limited scientific literature available during the development of the 2006 TMDL did not allow further quantification of mercury loads in urban stormwater.

As described in the TMDL Technical Support Document (Appendix A: Technical Support Document), current urban stormwater runoff volumes and erodible soil mercury concentrations were estimated to determine that the current MS4-permitted urban and highway stormwater

Oregon Department of Environmental Quality 51 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 areas contribute approximately 3.1 percent of the total mercury load to waterways in the basin. All other non-permitted urban areas contribute approximately 1 percent of the total mercury load to waterways in the basin. The estimated mercury loads from individual permitted MS4 jurisdictions range from 0 to 0.93 kg/yr (for a total At-source load of 5.0 kg/yr) and loads from all non-permitted urban stormwater areas combined is estimated at 0.92 kg/yr.

The potential contributions of atmospherically-deposited mercury from stormwater managed through all of the general stormwater permits covering industrial and construction activities (NPDES 1200-A, 1200-Z 1200-C, 1200-CA and 1200-CN) were implicit within these modeled loads from urban stormwater runoff. Therefore, potential mercury loads from both MS4 permits and most general stormwater permits are addressed with assignment of point source wasteload allocations in this report. In contrast, mercury loads from non-permitted urban stormwater areas and implicit loads from the general stormwater permits that are effective in those areas, are addressed through assignment of nonpoint source load allocations section of this report. 9.4.2.1 Municipal stormwater permits Federal regulations require qualified municipalities, such as cities, counties and special districts, to obtain NPDES permit coverage for their stormwater discharges. These permits are referred to as municipal stormwater permits or Municipal Separate Storm Sewer System permits, referred to throughout the document as MS4 permits. The Phase I MS4 permits regulate discharges from municipal separate storm sewer systems owned or operated by Oregon’s largest cities and counties, as well as the highway system managed by the Oregon Department of Transportation. The Phase II MS4 permits cover the next most populated parts of the state that have urbanized areas as defined by the US Census Bureau. In order to reduce pollutants from urban runoff entering waters, the permits establish conditions, prohibitions and management practices applicable to discharges of urban stormwater. These actions must be protective of water quality in streams that receive permitted discharges and must meet the requirements of the federal Clean Water Act.

DEQ identified 47 entities that require coverage under Phase I and Phase II MS4 permits within the Willamette Basin, as illustrated in Table 9-5 below. This includes the Portland metro area, the smaller metro areas of Eugene, Salem and Corvallis, as well as other, smaller cities. Permit requirements include development of stormwater management programs as well as requirements for reporting on implementation progress over time. For more information about MS4 permits, visit: https://www.oregon.gov/deq/wq/wqpermits/Pages/MS4-Permits.aspx

Table 9-5. Summary of NPDES MS4 Phase I and Phase II Jurisdictions in the Willamette Basin. DEQ Jurisdictions EPA EPA Permit Permit Type Permittee Permit Covered Class Number Number MS4 Phase I Multnomah County Multnomah County Major ORS120542 103004

Portland and City of Portland MS4 Phase I Major ORS108015 101314 Co-Applicants Port of Portland

City of Gresham MS4 Phase I Gresham, Fairview Major ORS108013 101315 City of Fairview MS4 Phase I Eugene City of Eugene Major ORS107989 101244 MS4 Phase I Salem City of Salem Major ORS108919 101513

Oregon Department of Environmental Quality 52 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

DEQ Jurisdictions EPA EPA Permit Permit Type Permittee Permit Covered Class Number Number Clean Water Services Washington County City of Banks City of Beaverton ORS108014

City of Cornelius City of Durham Clean Water MS4 Phase I Major 101309 Services City of Forest Grove

City of Hillsboro City of King City City of North Plains City of Sherwood City of Tigard City of Tualatin Clackamas County

City of Gladstone

City of Johnson City City of Lake Oswego City of Milwaukie

City of Oregon City City of West Linn Clackamas County MS4 Phase I Major ORS108016 101348 Group City of Wilsonville

Oak Lodge Sanitary District Water Environment Services City of Happy Valley

City of Rivergrove

Oregon ODOT Statewide MS4 Phase I Department of Minor ORS110870 101822 Transportation Transportation MS4 Phase II Keizer City of Keizer Minor ORS110870 101822 MS4 Phase II Philomath City of Philomath Minor ORS112241 102914 MS4 Phase II Wood Village City of Wood Village Minor ORS098909 102911 MS4 Phase II Benton County Benton County Minor ORS113609 102912 MS4 Phase II Lane County Lane County Minor ORS113606 102895 MS4 Phase II Marion County Marion County Minor ORS113608 102905 MS4 Phase II Polk County Polk County Minor ORS116224 102906 MS4 Phase II Springfield City of Springfield Minor ORS084048 102896

Oregon Department of Environmental Quality 53 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

DEQ Jurisdictions EPA EPA Permit Permit Type Permittee Permit Covered Class Number Number MS4 Phase II Corvallis City of Corvallis Minor ORS113605 102913 MS4 Phase II Turner City of Turner Minor ORS113607 102907 MS4 Phase II Linn County Linn County Minor ORS126417 33174 MS4 Phase II Millersburg City of Millersburg Minor None None MS4 Phase II Albany City of Albany Minor None None

9.4.2.2 Industrial Stormwater General Permits There are currently no individual industrial stormwater-only permits active within the Willamette Basin. DEQ requires certain industrial facilities that discharge stormwater either directly to waterbodies or through storm drain conveyances to obtain general industrial stormwater permit coverage, referred to as the 1200-Z permit. Examples of industrial activities that require permit coverage include manufacturing, transportation, mining, and steam electric power industries, as well as scrap yards, landfills, certain sewage treatment plants, and hazardous waste management facilities. DEQ also requires industrial facilities that discharge stormwater associated with sand and gravel mining activities to obtain industrial stormwater permit coverage, referred to as the 1200-A permit. The 1200-Z and 1200-A permits require that industrial facilities develop stormwater control plans, monitor for stormwater pollutants, and implement best management practices to reduce impacts on stormwater from industrial activities. Within the Willamette Basin, there are approximately 629 registrants under the 1200-Z permit and 109 registrants under the 1200-A permits. Most of these permits are within urbanized areas. For more information about the 1200-Z or 1200-A permit, visit: https://www.oregon.gov/deq/wq/wqpermits/Pages/Stormwater-Industrial.aspx

9.4.2.3 Construction Stormwater Permits Federal regulations require construction sites that disturb one acre or more to obtain NPDES permit coverage for their stormwater that discharges either directly to waterbodies or through a storm drain conveyance system. Construction sites that disturb less than one acre but are part of a common plan of development are also required to obtain permit coverage. Construction stormwater general permits are commonly referred to as either the 1200-C, 1200-CN or 1200- CA permit. The 1200-CN is applicable to construction projects performed within certain qualified local municipalities, and the 1200-CA is issued to public agencies. Construction activities addressed by the permit include clearing, grading, excavation, and materials or equipment staging and stockpiling.

The construction stormwater permits are designed to prevent the discharge of polluted stormwater from construction sites that can harm aquatic life and reduce water quality. These pollutants are commonly in the form of muddy or turbid stormwater runoff, which can transport debris and chemicals that come into contact with stormwater. Within the Willamette Basin, there are approximately 1,000 registrants under the 1200C/CN/CA permits, and these projects occur mostly within urbanized areas. Due to the ephemeral nature of construction activities, the number and location of 1200C/CN/CA registered projects will change over the life of the TMDL. For more information about the 1200-C, 1200-CN or 1200-CA permit, visit: https://www.oregon.gov/deq/wq/wqpermits/Pages/Stormwater-Construction.aspx

Oregon Department of Environmental Quality 54 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 10. Allocations

To meet water quality standards, loading from all pollutant sources must not exceed the loading capacity of a waterbody, including an appropriate margin of safety. Allocations are quantified measures that assure water quality standards will be met and may distribute the pollutant loads between nonpoint and point sources. Load allocations are portions of the loading capacity that are attributed to natural or background sources, such as soils, or from nonpoint sources, such as urban, rural agriculture, forestry activities or atmospheric sources. Wasteload allocations are portions of the total load that are assigned to point sources of pollution, such as sewage treatment plants or specific industries. The wasteload allocations are used to establish effluent limits in discharge permits. Allocations can also be reserved for future uses.

This TMDL is developed to restore and maintain water quality standards for mercury throughout the entire basin. The most restrictive of the Oregon numeric criteria for mercury is the fish tissue methylmercury criterion of 0.040 mg/kg. Bioaccumulation of methylmercury into fish tissue is a complex interaction of total mercury loading and methylation and demethylation processes in the aquatic system and food web dynamics. The allocations and surrogates described in this section were developed to reduce total mercury loading to the Willamette Basin and reduce methylation processes, which DEQ identified as necessary for meeting the fish tissue methylmercury criterion.

DEQ anticipates that implementation of allocations and surrogates to reduce high levels of solids and total mercury in the water column during periods of high flows will reduce the amount of mercury available for methylation, which in turn, will reduce the amount of methylmercury in the water. Lower levels of methylmercury in the water column will lead to less methylmercury bioaccumulation in fish tissue and eventually lower fish tissue methylmercury concentrations.

Using the linkages of mercury sources to Willamette Basin streams described in Section 6, this section assigns the mercury load allocations needed in order to achieve the mercury water quality criteria within the basin. Allocations are assigned to different sources of mercury; these sources are assessed in Section 9. The allocations are expressed as percent reductions and are based on the best estimates of total mercury loading. The complex behavior of mercury in the environment and the dominance of global, national and regional atmospheric deposition of mercury does not currently allow for distinctions between natural, other background (including long-range transport) and anthropogenic components of mercury sources. These source components exist in mixtures, both in modeled simulations and physically on the landscape. Therefore, the allocations that follow are as refined as possible within sectors, but are comprised of unquantified background and anthropogenic source elements, each with varying levels of potential control.

Nonpoint sources are the ones most affected by these mixtures of sources. These were not separated out to identify specific sources within the aggregated allocation. Rather, the broad category captures “atmospheric deposition” through the source categories described in the TMDL Technical Support Document as “sediment erosion,” “surface runoff” and “atmospheric deposition direct to streams.” The relative contributions of the different sources for the existing loads were estimated using the mass balance model and are listed in Table 10-1. DEQ’s expectation is that all applicable management strategies will be applied to the controllable portions of each source in order to achieve each responsible entity’s portion of the aggregated reductions needed. As noted in Section 6.1.5, the allocation approach is conservatively based on at-source, rather than delivered, loads at the basin-scale. This added conservatism reduces

Oregon Department of Environmental Quality 55 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

uncertainty and adds confidence that basin-scale allocations will be sufficient for water quality criteria to be met and address mercury impairments in all subbasins with implementation of measures to achieve the allocated reductions decribed below.

Table 10-1. Summary of TMDL Components. Mercury Water Quality 0.040 mg/kg fish tissue Criterion Total Mercury TMDL Water 0.14 ng/L Column Target Total Mercury Loading 43.36 g/day or 15.83 kg/year Capacity

SOURCE SECTORS EXISTING LOADS ALLOCATIONS Relative Relative Percent Allocation g/day kg/year Contribution g/day kg/year Reduction of Load to Total Load Capacity General Nonpoint

Source and

Background1

Captures: Forestry,

Agriculture, Water 124.82 88%3 29.68 10.84 68.46% Impoundments, Water 341.742 94.5%2 2 Conveyance Entities Non-Permitted Urban 75% 0.65 0.24 1.5% Stormwater

NONPOINT SOURCES Atmospheric 11% 5.73 1.96 12.38% Deposition Legacy Metals Mines 4.00 1.46 1.1% 95% 0.22 0.80 0.5%

NPDES Wastewater Point Source 4.44 1.62 1.2% 10% 4.12 1.50 9.5% Discharges NPDES Permitted Stormwater Point Source Discharges 11.31 4.13 3.2% 75% 2.90 1.06 6.7% POINT SOURCES Reserve Capacity Not Applicable 1%4 0.43 0.16 1.0% Margin of Safety Not Applicable Implicit TOTALS 361.49 132.03 100% NA 43.36 15.83 100% Notes: 1 Combines the following source categories from the TMDL Technical Support Document: Sediment Erosion, Surface Runoff, Groundwater, Atmospheric Deposition to Direct to Streams, Urban DMAs 2 Existing loads for General Nonpoint Source, Non-Permitted Urban Stormwater and Atmospheric Deposition were calculated in combination, though allocations for these three source categories are assigned separately. 3There is an additional 3.5% overall reduction from General Nonpoint Source and Background that results from the 11% decrease in Atmospheric Deposition, which reduces the mercury in precipitation that generates surface runoff. The additional reduction is calculated from the output of the Mass Balance Model. 4 Reserve Capacity is not allocated as a percent reduction, rather 1 percent of the allocated loading capacity will be reserved for any needed capacity for new or expanded point sources.

The components of the TMDL equation and its relationship to the load capacity are given below = + + +

𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 Σ𝑊𝑊𝑊𝑊𝑊𝑊𝑊𝑊 Σ𝐿𝐿𝐿𝐿𝐿𝐿 𝑀𝑀𝑀𝑀𝑀𝑀 𝑅𝑅𝑅𝑅 ≤ 𝐿𝐿𝐿𝐿

Oregon Department of Environmental Quality 56 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

is the sum of the waste load allocations, is the sum of the load allocations, is the margin of safety, is the reserve capacity, and is the load capacity. Σ𝑊𝑊𝑊𝑊𝑊𝑊𝑊𝑊 Σ𝐿𝐿𝐿𝐿𝐿𝐿 𝑀𝑀𝑀𝑀𝑀𝑀 The TMDL equation was𝑅𝑅𝑅𝑅 reorganized for clarity to incorporate𝐿𝐿𝐿𝐿 the basin-wide reserve capacity of 1%. The reserve capacity was set aside for future growth and is not available for the waste load allocations or load allocations. To demonstrate this we subtracted reserve capacity from both sides of the inequality of the TMDL equation, which resulted in the following form: = + + + = + + 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 Σ𝑊𝑊𝑊𝑊𝑊𝑊𝑊𝑊 Σ𝐿𝐿𝐿𝐿𝐿𝐿 𝑀𝑀𝑀𝑀𝑀𝑀 𝑅𝑅𝑅𝑅 − 𝑅𝑅𝑅𝑅 ≤ 𝐿𝐿𝐿𝐿 − 𝑅𝑅𝑅𝑅 The reserve capacity is subtracted𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 fromΣ𝑊𝑊𝑊𝑊 the𝑊𝑊𝑊𝑊 loadΣ𝐿𝐿𝐿𝐿𝐿𝐿 capa𝑀𝑀city𝑀𝑀𝑀𝑀 and≤ 𝐿𝐿𝐿𝐿 the− re𝑅𝑅𝑅𝑅maining load capacity is allocated among the waste load allocations and the load allocations. The remaining load capacity is: = 1% × = 99% ×

= 43.36 𝐿𝐿𝐿𝐿 −1%𝑅𝑅𝑅𝑅× 43𝐿𝐿𝐿𝐿.36− =𝐿𝐿𝐿𝐿43.36 𝐿𝐿𝐿𝐿0.43 = 42.93 𝑔𝑔 𝑔𝑔 𝑔𝑔 𝑔𝑔 𝑔𝑔 𝐿𝐿𝐿𝐿 − 𝑅𝑅𝑅𝑅 − − From the equation above, the𝑑𝑑𝑑𝑑𝑑𝑑 reserve capacity𝑑𝑑𝑑𝑑𝑑𝑑 is 0.43 g / day.𝑑𝑑𝑑𝑑𝑑𝑑 The remaining𝑑𝑑𝑑𝑑𝑑𝑑 compo𝑑𝑑𝑑𝑑𝑑𝑑nents of the TMDL equation using the eighth column of Table 10-1 are: = 9.5% × 43.36 + 6.7% × 43.36 = 7.02 𝑔𝑔 𝑔𝑔 𝑔𝑔 Σ𝑊𝑊𝑊𝑊𝑊𝑊𝑊𝑊 = 80.8% × 43.36 + 1.5%𝑑𝑑𝑑𝑑𝑑𝑑× 43.36 + 0.5%𝑑𝑑𝑑𝑑𝑑𝑑× 43.36 𝑑𝑑𝑑𝑑𝑑𝑑= 35.91 𝑔𝑔 𝑔𝑔 𝑔𝑔 𝑔𝑔 Σ𝐿𝐿𝐿𝐿𝐿𝐿 = + + + 𝑑𝑑𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑𝑑𝑑 g = 7.02 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇+ 35.91Σ𝑊𝑊𝑊𝑊𝑊𝑊𝑊𝑊+ Σ𝐿𝐿𝐿𝐿𝐿𝐿 𝑀𝑀+𝑀𝑀𝑀𝑀0.43 𝑅𝑅𝑅𝑅 = 43.36 day 𝑔𝑔 𝑔𝑔 𝑔𝑔 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖 is implicit and discussed in Section 11𝑑𝑑𝑑𝑑𝑑𝑑. 𝑑𝑑𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑𝑑𝑑 𝑀𝑀In 𝑀𝑀𝑀𝑀alignment with OAR 340-042-0040(6), DEQ considered several factors in distributing the allocations among the identified sources using a basin-wide approach. The factors considered include: • Atmospheric deposition is the major source of mercury throughout the Willamette Basin. • Atmospheric deposition generally follows precipitation patterns and does not result in specific “hotspots.” • Relative contributions of sources, including land area and discharge flow volumes. • Level of assurance that reductions needed from nonpoint and point sources are achievable with implementation of available management actions. • Maturity of programs and existing efforts.

DEQ’s allocation approach was informed by sources of mercury in the basin, sources of mercury excess loads and basin loading capacity, and the level of implementation of existing mercury control strategies. Point source discharges contribute a relatively small portion of the total load of mercury within the basin. Many of the largest municipal sewage treatment dischargers have been implementing measures that reduce mercury for many years and the technology for achieving additional reductions from those facilities is very limited. Reductions that could be achieved from the already small contribution from these sources are minimal and as a result, wastewater discharges received a relatively small reduction requirement.

Oregon Department of Environmental Quality 57 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

In contrast to point sources, nonpoint source land use activities are present over a significant portion of land area within the basin. These activities can mobilize to waters mercury that has been deposited on the landscape and bound to sediment. Aggregated together with background sources, the nonpoint source general category contributes approximately 94 percent of the total mercury load in the basin. Furthermore, the total mercury reduction potential from these sources is high because some sources in this category have not fully implemented mercury minimization measures. The large aggregated load means that even relatively small percentage reductions would achieve larger quantitative declines in loading. As a result, a large reduction requirement was applied for nonpoint sources. In the case of mining, the relative contribution is low, but the sources are discrete and isolated and there is a high potential for reduction upon remediation. Therefore, legacy metals mining received a very high reduction requirement. In the Coast Fork subbasin, this 95 percent legacy mining reduction, taken together with the 88 percent nonpoint source reduction and any point source reductions (stormwater 75 percent and wastewater 10 percent), amounts to an effective total reduction of approximately 93 percent. Finally, in the case of stormwater, DEQ assigned a large reduction, despite the relative contribution being low. Support for this approach includes the fact that the affected jurisdictions cover large land areas within the basin that collect relatively large portions of atmospherically deposited mercury. In addition, effective mercury minimization measures are feasible for achieving reductions. The level of maturity of stormwater control programs ranges from highly sophisticated for long- standing permitted jurisdictions to smaller communities that have not yet begun fully implementing control measures. Along with relative contributions by source types, the reductions needed and the resultant distribution of the load capacity among the sources are summarized above in Table 10-1 and explained in greater detail by category in the sections that follow. For more information, please refer to the allocation spreadsheet, which calculates wasteload and load allocations for the Willamette Basin Mercury TMDL, found in Appendix B. The analysis presented in this document supports DEQ’s conclusion that no one source category is entirely responsible for the mercury contamination in the Willamette Basin. Collaborative efforts extending across all point source and nonpoint source categories will be necessary to achieve reductions in mercury loading and, ultimately, the restoration of the beneficial use of fish consumption. DEQ anticipates development and implementation of effective mercury minimization measures and other management strategies appropriate to individual facilities and land activities as the primary control mechanisms for achieving reductions in the controllable components of each sector’s source load. A description of the various implementation activities and management strategies designed to achieve cross-sector reductions in the load of total mercury are presented in detail in the Water Quality Management Plan. 10.1 Load allocations for nonpoint sources Load Allocations OAR 340-042-0040(4)(h), 40 CFR 130.2(g): This element determines the portions of the receiving water's loading capacity that are allocated to existing nonpoint sources including background sources. The mercury load allocations in the Willamette Basin is a mixture of background loads and anthropogenic nonpoint sources.

Oregon Department of Environmental Quality 58 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

As summarized in Table 10-1 above, load allocations for Nonpoint Source Sectors are: • General Nonpoint Source Sector: 88 percent reduction of grams of total mercury per day and includes the following categories: o Forestry o Agriculture o Water Impoundments o Water Conveyance entities o County-owned lands, properties, facilities and roads (outside of MS4 permit jurisdictional areas) o Background sources (groundwater and atmospheric deposition) • Legacy Metals Mining Sector: 95 percent reduction of grams of total mercury per day • Non-Permitted Urban Stormwater Sector: 75 percent reduction of grams of total mercury per day (applies to cities not regulated under an MS4 permit) • Atmospheric Deposition: 11 percent reduction of grams of total mercury per day

A single reduction percentage is being used for the General Nonpoint Source Sector load allocation, because current analysis does not enable DEQ to separate background and anthropogenic sources. As noted above in Table 10-1, the reduction of 88 percent for the load allocation will be applied to all nonpoint sources except legacy metals mining, atmospheric deposition and stormwater in developed city areas that are not covered by an MS4 permit. Contributions to the overall mercury load from these developed city areas were estimated using the jurisdictional boundaries of these communities. Most of the mercury load from this source is from atmospheric deposition, but controlling surface runoff and soil erosion will reduce the mercury loads from these developed areas from entering rivers and streams in the basin. The same is true for all of the other nonpoint sources. The 88 percent reduction for anthropogenic sources will be addressed using runoff and erosion control approaches. The 11 percent reduction for atmospheric sources is anticipated to occur through controls on local emissions within Oregon, but to greater extent through on-going reductions being achieved nationally (United Nations Environment Programme, 2019) and in the future through enactment and implementation of international treaties. This reduction in atmospheric sources is well below the reduction used in approved mercury TMDLs throughout the US, which range from 67 percent to 90 percent (Limno Tech, 2018; North Carolina Department of the Environment and Natural Resources, 2012; Minnesota Pollution Control Agency, 2007; New England Interstate Water Pollution Control Commission, 2007). DEQ’s expectation is that all relevant management strategies will be applied to the controllable portions of each source toward achieving each responsible entity’s portion of the aggregated Nonpoint Source General Sector reductions needed. 10.2 Wasteload allocations for point sources OAR 340-042-0040(4)(g), 40 CFR 130.2(h) This section describes the portions of the Willamette Basin’s loading capacity that are allocated to existing point sources of pollution, including all point source discharges regulated under the Federal Water Pollution Control Act Section 402 (33 USC Section 1342).

As summarized in Table 10-1 above, wasteload allocations for NPDES permitted sources are: • Wastewater dischargers: 10 percent reduction of grams of total mercury per day and includes the following permit categories: o Major and minor domestic Sewage Treatment Plant wastewater permits o Major and minor Industrial wastewater permits o Wastewater discharges covered under General permits

Oregon Department of Environmental Quality 59 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

• Stormwater dischargers: 75 percent reduction of grams of total mercury per day and includes the following permit categories: o Municipal Separate Storm Sewer System – Phase I o Municipal Separate Storm Sewer System – Phase II (general and individual) o Industrial Stormwater (1200-A and 1200-Z general permits) o Construction Stormwater (1200-C/CN/CA general permits)

The 2006 Willamette Basin Mercury TMDL assigned interim waste load allocations to the existing point source discharges at the time, with the expectation that point source mercury load reductions would be achieved through implementation of mercury minimization measures. WLAs were assigned in two categories, as follows: Publicly Owned Treatment Works Discharges at 2.6 kg/yr (or about a 26 percent reduction) and Industrial Discharges at 1.1 kg/yr (or about a 27 percent reduction).

In the 2006 TMDL, urban stormwater discharges were not distinguished from nonpoint source runoff and erosion. While the 2006 TMDL acknowledged that MS4-permitted stormwater discharges are considered point sources, load reductions from both permitted and unpermitted stormwater were contained within the ‘runoff of atmospherically deposited mercury’ and ‘erosion of mercury containing soils’ categories and accounted for in interim nonpoint source load allocations amounting to about 27 percent reductions from both categories.

As noted in Section 9.4 on point source assessment, more data and information on both effluent flow and concentration is now available, which allowed more accurate and refined loading estimates of permitted municipal and industrial wastewater discharges and municipal stormwater discharges.

In development of this TMDL, DEQ considered whether individual waste load allocations for point sources were appropriate. DEQ completed revised evaluations under stricter and more demanding scientific standards requiring analysis of significantly more data and new information on factors affecting mercury pollution, including multiple potential sources, bioaccumulation patterns and changes in the types of mercury being released and transformed in the large and complex Willamette Basin. These revised evaluations indicate that the estimated mercury load from permitted municipal and industrial wastewater point sources is significantly lower (1.1 percent of total load) than was estimated in the 2006 evaluation (3.9 percent). As discussed in the TMDL Technical Support Document, deposition of mercury onto the Oregon landscape is the dominant source of mercury reaching Willamette Basin streams. While these deposited air emissions originate as a mix of global, national, regional and local sources, the largest portion is derived from historical deposition of global anthropogenic mercury emissions (TetraTech, 2019a), or background sources outside of DEQ’s control, per Oregon’s definition in OAR 340- 042-0030. Further, mercury loads from all permitted (wastewater and stormwater) point source discharges combined are conservatively estimated to be approximately four percent of the total load to Willamette Basin streams. As was found in the 2006 TMDL analysis, even total elimination of this estimated 1.1 percent wastewater and the 3 percent estimated municipal stormwater contributions would not result in measurable response in terms of lowered mercury in the streams, due to the far greater proportion of contributions from atmospheric deposition and nonpoint source delivery to streams, as well as the decades long lag time for measureable in-stream response. However, DEQ recognizes that, as an environmentally persistent bioaccumulative toxic substance, mercury should be eliminated from discharges to the extent practicable. Therefore, based on the Clean Water Act’s allowance for aggregate or individual allocations (40 CFR 130.2(i)); EPA’s Guidance for implementing the January 2001 Methylmercury WQ Criterion (2010) and EPA’s Memo on Elements of Mercury TMDLs Where Mercury Loadings are Predominantly from Air Deposition (2008); precedents of EPA approved

Oregon Department of Environmental Quality 60 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

mercury TMDLs of 21 other states (dated 2001-2018); and as indicated by a rigorous scientific evaluation, DEQ is assigning aggregate wasteload allocations for municipal and industrial wastewater and municipal stormwater point source discharges. The wasteload allocations that follow meet the intent of individual allocations by requiring site-specific permit requirements and monitoring with enforceable conditions, such that individual site reductions will be completed and will cumulatively add up to the aggregate percent reduction requirements by sector set by the TMDL.

The wasteload allocation implementation approach specific to each sector and permit category is presented in detail in the Water Quality Management Plan. A summary of DEQ’s expectation for implementation of wasteload allocations for the two NPDES-permitted sectors is as follows:

1. NPDES-permitted municipal and industrial wastewater discharges must implement mercury minimization measures, as warranted by mercury monitoring results, toward achieving a cumulative 10 percent reduction of the existing estimated wastewater point source mercury load into Willamette Basin streams.

2. NPDES-permitted MS4 stormwater jurisdictions must implement mercury and erosion minimization measures, with quantifiable objectives, toward achieving an appropriate portion of the needed cumulative 75 percent reduction of the estimated existing mercury load, from mixed atmospheric deposition and human-caused disturbance, transported in stormwater runoff to Willamette Basin streams.

For wastewater discharges, DEQ will require development and implementation of mercury minimization programs at facilities with measured mercury, significant flows and activities that increase the potential for mercury in discharge, in order to achieve facility-specific portions of the aggregate 10 percent overall sector reduction. The facility-specific portions will reflect both current minimization programs and the potential for reductions from current conditions. This approach is consistent with EPA methylmercury implementation guidance, which urges source reduction over other approaches (U.S. Environmental Protection Agency, 2010a) Available data and information from Oregon, California, Minnesota and Michigan demonstrate achievement of mercury load reductions from municipal sewage treatment discharges using mercury minimization programs. DEQ found, during evaluation and development of potential mercury variances within the Willamette Basin, that no feasible treatment technologies have been demonstrated at the scale of major sewage treatment plants and large industrial facilities to achieve mercury effluent concentrations below 1 ng/L to 3 ng/L. Advanced wastewater treatment facilities have achieved effluent concentrations between 1 ng/L and 3 ng/L in some cases, while others have achieved levels up to 5 ng/L. Facilities with secondary treatment technologies that have implemented mercury minimization programs for a decade or more, have also been able to achieve mercury effluent concentrations of 3 ng/L to 5 ng/L. Therefore, enhancing and implementing mercury minimization programs is anticipated to be the most effective approach to achieve aggregate reductions for this sector. Please see Appendix C: Variance justification excerpts for supporting information.

DEQ determined that flows from minor municipal wastewater discharges and all general wastewater permit categories were very minor with regard to mercury loading from the overall wastewater sector into Willamette Basin streams. However, as data and information warrant, some minor industrial permittees with activities that may increase the potential for mercury in their discharge may be required to implement Mercury Minimization Programs.

Also within the aggregated wastewater sector, DEQ is proposing to prohibit discharges from suction dredges under the General NPDES 700PM permit in streams with known mercury

Oregon Department of Environmental Quality 61 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

contamination from historical mercury and gold mining activities. Studies in Oregon, California, Wisconsin and Florida have shown that mercury in stream beds is disturbed, mobilized and methylated by suction dredging (Fleck, et al., 2010; Gray, Hines, Krabbenhoft, & Thoms, 2012; Humphreys, 2005; Marvin-DePasquale, et al., 2009; Marvin-DiPasquale, et al., 2011). Soils and stream sediment sampling in the former Bohemia Mining District indicates high concentrations of mercury. Mercury concentrations found in stream-side soils range from 13 mg/kg to >50 mg/kg and stream sediments in Brice Creek, Champion Creek, Sharps Creek and the Row River upstream of Dorena Reservoir range from 0.08 mg/kg to 8.78 mg/kg (Hygelund, Ambers, & Ambers, 2001) and (Tobias & Wasley, 2013). These streams are tributary to the Dorena Reservoir, which is 303(d) listed for mercury and has fish advisories for mercury contamination in place. Therefore, upon renewal of the 700PM permit, DEQ will prohibit suction dredge mining in locations in streams that flow from the former Bohemia Mining District and are tributary to the Dorena Reservoir (including Row River, Brice Creek, Sharps Creek, and Champion Creek). While suction dredge disturbance of mercury laden sediment in these streams is currently intermittent and releases and methylation potential are not quantifiable, these prohibitions in this known historical source area will add to reductions achieved throughout the basin toward the 10 percent aggregated WLA for the wastewater sector.

Because the 10 percent overall reduction amounts to reducing the current load from aggregated wastewater discharges of 4.44 g/day to 4.00 g/day, DEQ anticipates that tracking of effectiveness of mercury minimization plan implementation will demonstrate achievement of this needed reduction. Permit requirements and timelines for implementation are described in the Water Quality Management Plan in Sections 13.3.2 and 13.4.2 of this document and the accountability framework for reasonable assurance that TMDL goals will be met is described in Section 14.1.

For municipal stormwater discharges, EPA guidance (U.S. Environmental Protection Agency, 2002; U.S. Environmental Protection Agency, 2008a; U.S. Environmental Protection Agency, 2008b; U.S. Environmental Protection Agency, 2010a; U.S. Environmental Protection Agency, 2010b; U.S. Environmental Protection Agency, 2014; U.S. Environmental Protection Agency, 2015; U.S. Environmental Protection Agency, 2016) and precedents in other states indicates that development and enhancement of measures targeted at mercury minimization and erosion control is the most effective and practical approach for achieving needed mercury reductions in stormwater discharges. As permits are renewed, DEQ will require reporting of measurable objectives for implementation of mercury and erosion controls across MS4 jurisdictions to achieve their potential portion of the 75 percent aggregate reduction. The mercury loads in MS4 discharges estimated in the modeling, which covered multiple years of collected and estimated data, were 11 g/day. As a result, the 75 percent aggregate reduction needed leaves 2.75 g/day of mercury allowable in these discharges. DEQ acknowledges that atmospheric deposition of mercury is the dominant driver of stormwater mercury loads and that MS4 controls will capture unknown quantities of both human-caused and background sources of mercury, which will not be distinguishable. Reductions in atmospheric deposition are anticipated to be reflected in reductions within MS4 jurisdictions. Because mercury loads from general stormwater permits (NPDES 1200Z, 1200A and 1200C/CN/CA) were implicit in the MS4 modeled load estimates, DEQ anticipates reductions achieved through erosion control requirements and reduced total suspended solids benchmarks for key geographic areas (for example, industrial stormwater discharges to Portland Harbor and the Columbia Slough) will also contribute to reductions needed in the overall stormwater sector wasteload allocation. DEQ also acknowledges that effective minimization measures by many MS4 jurisdictions have been ongoing for several years and reductions achieved are not reflected in the current modeling. DEQ anticipates that some portion of the needed reduction has already been achieved. As noted in the Water Quality Management Plan in Sections 13.3.2 and 13.4.2, permits will require reporting of evaluations of

Oregon Department of Environmental Quality 62 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

attainment of wasteload allocations and annual and five year intervals. DEQ will use this information in tracking and evaluating overall allocation attainment, as described in the accountability framework for reasonable assurance that TMDL goals will be met is described in Section 14.1. 10.3 Instream surrogate targets DEQ may use surrogate measures to estimate allocations for pollutants addressed in the TMDL (OAR340-042-0040(5)(b). DEQ may use one or more surrogate measures for a pollutant that is difficult to measure or highly variable. Typically, a surrogate measure will be closely related to the pollutant, and may be easier to monitor and track. The TMDL establishes the relationship between the surrogate measure and pollutant.

DEQ anticipates that implementation of allocations and surrogates to reduce high levels of solids and total mercury in the water column during high flows will reduce the amount of mercury available for methylation, which in turn, will reduce the amount of methylmercury in the water and bioavailability for uptake into fish. Lower levels of methylmercury in the water column will lead to less methylmercury bioaccumulated into fish tissue and eventually lower fish tissue methylmercury concentrations.

Total suspended solids, commonly referred to as TSS, is often used as a surrogate for pollutants A strong relationship was found between TSS and total mercury in urban catchments (Eckley & Branfireun, 2009). In addition, TSS was used as a surrogate for DDT (Dichlorodiphenyltrichloroethane) to meet instream targets for the Lower Yakima TMDL in Washington (Johnson, 2005). TSS can be a cost effective surrogate for total mercury in the water column which can be expensive to measure. A positive correlation between TSS and total mercury makes TSS a useful surrogate measure due to the capacity of total mercury to bind to particulate matter (Eckley & Branfireun, 2009). Therefore, provided that there is correlation between TSS and total mercury within the mainstem Willamette River and its tributaries, TSS could be used to reduce total mercury in stream and evaluate progress towards achieving the allocations and total mercury TMDL water column target described in this section. DEQ evaluated the relationship between TSS and total mercury for the Willamette Basin. A general description of the approach and results of the evaluation is described below. A detailed description of the approach and results is provided in Table 10-2. Based on the results of this analysis, TSS surrogate instream targets were assigned to supplement, but not to supplant, the allocations in this TMDL (Table 10-1).

All data were extracted from the Willamette Basin mercury database from January 1, 2002 to present. There were 63 paired samples used in this analysis from nine different HUC 8 subbasins within the Willamette Basin. All of the 63 surrogate pairs had mercury concentrations with a detection limit of 0.5 ng/L (U.S. Environmental Protection Agency, 2002) which is above the instream target of 0.14 ng/L total mercury set for the Willamette Basin. All of the samples used for the analysis were above this detection limit. The sites (sampling locations) used in the analysis were defined by the latitude and longitude coordinates where the samples were collected.

A Linear Mixed Effects model with sites (sampling locations) was used to calculate TSS concentrations that correspond to total mercury concentrations and was used to estimate TSS concentrations and calculate percent reductions of total mercury concentrations (Table 10-2). A strong relationship was found between TSS and total mercury for this data. Therefore, in addition to the load allocations and waste load allocations in Sections 10.1 and 10.2, the data

Oregon Department of Environmental Quality 63 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

analysis of instream total mercury and TSS was used to develop instream surrogate targets for TSS.

Based on the relationship found between total suspended solids and total mercury, surrogate instream targets were set for reductions in high levels of TSS concentrations to reduce total mercury in stream and evaluate progress towards achieving the allocations and total mercury TMDL water column target described in this section. These reductions of TSS are expected to reduce total mercury loads that could occur during high precipitation events and high flows. Instream surrogate targets were developed to set reductions in the 95-percentile for the TSS concentration over time. The initial 75 percentile of TSS concentration is based on the 2019 dataset used to develop the TSS and total mercury concentrations (Figure G-16). The schedule is listed in Table 10-2. The empirical cumulative distribution function will be reanalyzed when new data becomes available and a Bayesian inference approach to structured decision making (Conroy & Peterson, 2013) will be used to update the statistical distribution used to assess the progress in meeting the reduction targets and to reassess past targets if any had occurred.

Table 10-2. Schedule for instream mainstem Willamette and HUC8 outlets surrogate targets for reducing high total suspended solid concentrations during times of high precipitation events and high flows. Years Reduction in 95th Cumulative total Target for Maximum from Percentile of TSS reduction in TSS Instream TSS (mg/L) 2019 Concentration (mg/L) 0 0% 0 17 5 10% 1.7 15 10 25% 4.3 13 20 50% 8.5 9 30 75% 12.7 4 Notes: Based on the 95th Percentile of TSS Concentrations from the Dataset of 63 Surrogate Pairs. The TSS maximum concentration at the 95th percentile is 17 mg/L. Limits of high precipitation events and high flows are defined using EPA’s guidance for load duration curve analysis (U.S. Environmental Protection Agency, 2007).

The TSS surrogate targets will apply to the mainstem Willamette and HUC8 outlets. The TSS surrogate targets will be used for reducing total mercury instream and as one tool for evaluating progress towards achieving allocations and the total mercury TMDL water column target described in this section. In addition, because TSS is a cost effective surrogate it will be used to supplement but not supplant the allocations and TMDL water column target for evaluating TMDL implementation effectiveness. Locations will be described in the Assessment and Monitoring Strategy to Support Implementation of Mercury Total Maximum Daily Loads for the Willamette Basin and initial surrogate targets may be updated as described in the Accountability Framework (Section 14). Other surrogates for total mercury may be evaluated from the data collected as part of the Assessment and Monitoring Strategy.

Oregon Department of Environmental Quality 64 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 11. Margin of safety

OAR 340-042-0040(4)(1) The Clean Water Act requires that each TMDL be established with a margin of safety to account for uncertainty in available data or in the actual effect controls will have on loading reductions and receiving water quality. A margin of safety is expressed as unallocated assimilative capacity or conservative analytical assumptions used in establishing the TMDL (i.e., derivation of numeric targets, modeling assumptions or effectiveness of proposed management actions).

A margin of safety may be implicit through the use of conservative assumptions that result in more protective loading capacity, wasteload allocations, or load allocations. The margin of safety may also be explicitly stated as an added, separate quantity in the TMDL calculation. In any case, assumptions should be stated and the basis behind the margin of safety documented. The margin of safety is not meant to compensate for a failure to consider known sources.

Due to the complexity of the TMDL analysis, DEQ determined that an implicit margin of safety was appropriate. An implicit margin of safety was selected based on the following components of the TMDL analysis. DEQ defined the target surface water total mercury concentration based on the Northern Pikeminnow. Because the Northern Pikeminnow is the most efficient mercury bioaccumulator among the species considered due to its high tropic level, this provides a margin of safety. Calculating reductions based on the next most conservative species (Largemouth Bass) results in an 82 percent reduction needed in the basin (rather than 88 percent). However, this would not provide a margin of safety because this would not be protective of Northern Pikeminnow, one of the species for which there are currently fish consumption advisories in place. Another conservative aspect that provides a margin of safety is use of the median concentration, 0.14 ng/L of total mercury, to meet the fish tissue criterion (0.04 mg/kg methylmercury) in Northern Pikeminnow. The statistical distribution for the Monte Carlo simulations of the Food Web Model was right-skewed and use of the median as a measure of central tendency results in a lower value than the mean concentration. Using the mean concentration (0.23 ng/L) results in basin wide reductions needed of 81 percent, rather than 88 percent. However, using the mean-derived surface water total mercury target is predicted to result in exceedances of the fish tissue criterion for the Northern Pikeminnow, so does not provide a margin of safety. Another component of the margin of safety is that the Northern Pikeminnow is not a popular commercial or recreational target. Instead, pikeminnow may be consumed on an occasional basis by recreational or subsistance fishermen (related to the fish consumption rate of the water quality standard), which adds another margin of safety. Another conservative component in the TMDL analysis was the use of the total mercury concentration in fish tissue rather than methylmercury. The total mercury in fish is composed of 95 percent or greater methylmercury in higher trophic level piscivores (USEPA, 2000). Therefore using the total mercury concertation in fish tissue rather than methylmercury increases the margin of safety because the methylmercury concentration will be slightly less than the total mercury concentration. DEQ concluded that the conservative nature of these components of the TMDL analysis provide a sufficient margin of safety.

Oregon Department of Environmental Quality 65 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 12. Reserve capacity

As described in OAR 340-042-0040(k), reserve capacity is an element of the TMDL, which is an allocation for increases in pollutant loads from future growth and new or expanded sources. DEQ used an explicit reserve capacity of 1 percent. Reserve capacity may be granted by DEQ to NPDES permitted point sources. Prior to allocating a portion of the reserve capacity to a new or expanded point source, DEQ will require demonstration of effluent condition and implementation of DEQ approved mercury minimization measures, as described in the Water Quality Management Plan for the appropriate sector.

Oregon Department of Environmental Quality 66 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 13. Water Quality Management Plan

13.1 Introduction DEQ developed a Willamette Basin WQMP in 2006 that included pollutant reduction strategies for temperature, bacteria and mercury. Subbasin TMDLs within the Willamette Basin also included other pollutants. The WQMP associated with the TMDL at that time required stormwater control strategies for cities to reduce mercury and bacteria. Non-urban control measures were generally addressed through existing state and federal regulations and plans. To meet the mercury target established at that time required a 27% reduction of mercury. This updated WQMP replaces requirements established in 2006 for mercury. All other requirements for pollutants identified in the 2006 Willamette Basin WQMP remain effective.

This updated WQMP provides the framework for describing management efforts that will be put into action to attain the Willamette Basin Mercury Total Maximum Daily Load based on a more stringent fish tissue criterion that was adopted in 2011. This framework builds upon existing point and nonpoint source implementation plans to outline a management approach for reducing mercury from all land uses in the basin.

Oregon Administrative Rules (OAR 340-042-0040(4)(I)(G)) require DEQ to identify Figure 13-1. McKenzie River persons, including Designated Management Agencies that are responsible for implementing management strategies and sector-specific or source-specific implementation plans. A DMA is “a federal, state or local governmental agency that has legal authority of a sector or source contributing pollutants, and is identified as such by the Department of Environmental Quality in a TMDL” (OAR 340-042-0030(2)). See a complete list of DMAs and responsible persons in Appendix E: List of designated management agencies and responsible persons.

The WQMP includes a description of activities, programs, legal authorities and other measures for which DEQ and DMAs have regulatory authority. The WQMP also includes a description of how other responsible persons are expected to implement activities and programs that will help to achieve the TMDL.

TMDL modeling indicated that the major pathway for mercury entering waters of the state is through runoff and erosion of sediment containing mercury. Therefore, DEQ’s overall approach to meeting the fish tissue criterion is by reducing sources of mercury through runoff and erosion control management strategies, thereby reducing the total reservoir of mercury available to be methylated, which is the form that bioaccumulates in fish and shellfish. DMAs are encouraged to identify opportunities to better characterize conditions that may lead to methylation of mercury

Oregon Department of Environmental Quality 67 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 given that those processes are still an evolving science and that level of analysis was not available for this TMDL. 13.1.1 Implementation plans Following the issuance of a TMDL and WQMP, DEQ requires most DMAs and responsible persons to develop implementation plans that identify specific management strategies and actions that will be implemented in order to meet water quality standards over time. For DMAs and responsible When a nonpoint source DMA or responsible person persons associated with nonpoint complies with a DEQ approved implementation plan, sources of pollutants, these they are in compliance with TMDL requirements. implementation plans may be called different names. For example, implementation plans for the Bureau of Land Management and the U.S. Forest Service are called Water Quality Restoration Plans. The Oregon Department of Agriculture uses Agricultural Water Quality Management Area Plans to meet most requirements of an implementation plan.

According to regulations in OAR 340-042-0040(4)(l)(I), the WQMP must provide a schedule for submittal of implementation plans. DEQ typically gives DMAs and responsible persons 18 months to submit new or updated implementation plans following the issuance of a TMDL and WQMP. For this WQMP, DEQ will continue using the 18-month time frame for implementation plan submittal. Implementation plans must be posted to a publicly accessible website, unless the DMA does not have a website. DEQ reviews the plans in accordance with regulations in OAR 340-042-0080(4):

(a) Prepare an implementation plan and submit the plan to the Department for review and approval according to the schedule specified in the WQMP. The implementation plan must:

A. Identify the management strategies the DMA or other responsible person will use to achieve load allocation and reduce pollutant loading; a. Provide a timeline for implementing management strategies and a schedule for completing measurable milestones; b. Provide for performance monitoring with a plan for periodic review and revision of the implementation plan; c. To the extent required by Oregon Revised Statute 197.180 and OAR chapter 340, division 18, provide evidence of compliance with applicable statewide land use requirements, and; d. Provide any other analyses or information specified in the WQMP.

(b) Implement and revise the plan as needed.

In addition, implementation plans must provide an estimate of the technical and financial resources needed, associated costs, and the sources and authorities that will be relied upon to implement the plan. For more information, see Section 13.7 Costs and Funding.

For point sources, management strategies identified in the TMDL and WQMP Section 13.3.2 will be incorporated into renewed NPDES permits as enforceable provisions.

Following the issuance of the TMDL, DEQ may make a determination that nonpoint source implementation plans are not necessary for certain DMAs and responsible persons. In those

Oregon Department of Environmental Quality 68 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 cases, DEQ will provide a written determination to the DMA or responsible person of why a plan is not necessary. This determination could be based on factors, such as inaccurate identification of a DMA or responsible person, or documentation that an entity does not discharge to a waterbody. 13.1.2 Adaptive management The federal Clean Water Act and associated Oregon laws and implementing regulations governing water quality require water quality standards to be met over time. In some cases, responsibility may depend on practicability, but DEQ typically requires that all feasible steps be taken toward achieving the highest quality water attainable. This is a long-term goal in many watersheds, particularly where nonpoint sources of pollution are the main concern and significant landscape alterations are needed.

TMDLs are numerical allocations of pollutants that are set so that instream water quality standards are met. This TMDL includes values calculated from mathematical models and other analytical techniques designed to simulate and/or predict very complex physical, chemical and biological processes of mercury release and transport in the Willamette Basin. DEQ used models and techniques that incorporate large amounts of water quality and land use data and information specific to the Willamette Basin. However, in order to evaluate these processes on this scale, the models and techniques used simplify these complex processes and inherently contain a distribution of uncertainty concerning how streams and other waterbodies will respond to various management measures. For this Adaptive management is a process that reason, the TMDL is required to acknowledges and incorporates improved contain a margin of safety. technologies and practices over time in order to refine implementation. WQMPs are plans designed to reduce pollutant loads to meet TMDLs. DEQ recognizes that it will take time before management practices identified in a WQMP are fully implemented and effective in reducing and controlling pollution. In addition, DEQ recognizes that technology and practices for controlling nonpoint source pollution will continue to develop and improve over time. As implementation, technology and knowledge about these approaches progress, DEQ will use adaptive management to refine implementation. Figure 13-2 provides a conceptual representation of adaptive management.

Oregon Department of Environmental Quality 69 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Figure 13-2. Conceptual representation of adaptive management. The estimated timeline for achieving water quality standards is multiple decades.

DEQ also recognizes that despite best efforts, natural events beyond the control of humans may interfere with or delay attainment of the TMDL. Such events include, but are not limited to, floods, fire, insect infestations, and drought.

If a source is not given an allocation, it does not necessarily mean that a source is prohibited from discharging any wastes. DEQ may permit a point source that is not covered by an allocation to discharge if the holder either can adequately demonstrate that the discharge will not impact the pollutant in question, or that the discharge is covered by reserve capacity. If a nonpoint source DMA or responsible person complies with a DEQ-approved implementation plan, DEQ will consider them in compliance with the TMDL. DEQ has the following general expectations and intentions for using an adaptive management approach for the TMDL and WQMP: • Every five years, DEQ will review the progress of the TMDL and the WQMP. Where DEQ determines that implementation plans or effectiveness of management strategies are inadequate, DEQ will require DMAs and responsible persons to revise the components of their implementation plans to address these deficiencies. • In conducting this review, DEQ will evaluate the progress towards achieving the TMDL and water quality standards and the success of implementing the WQMP. • DEQ expects that each DMA and responsible person will also monitor and document its progress in implementing the provisions of its implementation plan. This information will be provided to DEQ for its use in reviewing the TMDL. This information is typically

Oregon Department of Environmental Quality 70 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

provided in an annual report and/or five year review report. Please see section 13.4 for more information on annual reporting and the five year review.

If DEQ determines that all appropriate measures are being taken by DMAs and responsible persons and water quality standards will still not be met, DEQ may take one of several actions depending on the information available. For example, DEQ may conduct a use attainability analysis if the current designated beneficial use of a waterbody cannot be met. In addition, DEQ may also consider reopening and modifying the TMDL, subject to available resources, if new information showed that the TMDL or associated surrogates should be modified. 13.2 Elements of the Water Quality Management Plan OAR 340-042-0040(4)(l) describes WQMP requirements. This section provides the framework of management strategies to attain and maintain water quality standards. The framework is designed to work in conjunction with detailed plans and analyses provided in sector-specific or source-specific implementation plans. Additional detail on each element is provided in the sections that follow. 13.2.1 Condition assessment and problem description As noted in OAR 340-042-0040(4)(l)(A), WQMPs must contain an assessment of conditions and description of the problem the TMDL is developed to address. Fish tissue and water samples were collected from the Willamette Basin and analyzed for mercury. The data indicated several segments of the Willamette River and its tributaries are not meeting water quality standards. Based on Oregon’s assessment methodology for the Integrated Report these waterbodies were identified as impaired and included on the state’s 303(d) list. Oregon’s 2012 Integrated Report contains the most recent listings relative to mercury for the Willamette Basin.

The Oregon Health Authority is responsible for evaluating contaminant concentrations in fish tissue, calculating the number of meals per month that can safely be consumed, and providing that information to the public by issuing a fish consumption advisory when data are available. DEQ helps to support this process by collecting and analyzing fish tissue samples and sharing these data with OHA. EPA and the National Parks Service also provide fish tissue data to OHA.

Advisories are designed to protect the public from contaminants sometimes found in fish, while also balancing the positive health benefits from eating fish. Information regarding fish consumption advisories can be accessed on OHA’s website: https://www.oregon.gov/oha/pages/index.aspx.

There are multiple fish consumption advisories for the Willamette Basin advising people of the health risks associated with consuming fish containing elevated levels of mercury. Currently, fish consumption advisories in place for mercury include: • Bass in all Oregon waters; • All resident fish (except stocked, fin-clipped rainbow trout 12-inches or less) in the Dorena and Cottage Grove Reservoirs; and • Resident fish in the mainstem Willamette River from its mouth on the Columbia River southward to Eugene, including the Coast Fork Willamette up to the Cottage Grove Reservoir.

These fish consumption advisories for mercury in the Willamette Basin and several 303(d) listings for mercury impaired waters support the need for additional mercury reductions in order

Oregon Department of Environmental Quality 71 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

to restore the beneficial use of “fishing”, and being able to safely eat fish. Table 3-1 contains the 303(d) listed waterbodies addressed in this TMDL.

The TMDL and accompanying WQMP demonstrate how Oregon will meet standards for total mercury in water and methylmercury in fish tissue, as well as the narrative water quality standard for toxic pollutants. The fish tissue methylmercury standard is 0.040 milligrams methylmercury/kilogram of fish tissue. 13.2.2 Goals and objectives Another required component of the WQMP is a section on goals and objectives, as described in OAR 340-042-0040(4)(l)(B). The overarching goal of this WQMP is to achieve the water quality standards for mercury in the Willamette Basin over time. Oregon has a mercury water quality standard to protect aquatic life, a methylmercury standard measured in fish tissue, and a narrative water quality standard for toxic chemicals (Section 13.2.1). The fish tissue standard, if not exceeded, protects those who consume up to approximately 23 eight-ounce servings of fish or shellfish every month from Oregon lakes and streams.

The primary objective of this WQMP is to lay out a framework that describes who is responsible for implementing the TMDL, management efforts that will be put into action in order to meet the TMDL, and how to measure progress towards attaining water quality standards for mercury.

The management strategies necessary to meet the TMDL load and wasteload allocations differ based upon the source of pollution and the responsibilities and resources of DMAs and responsible persons. Many DMAs and responsible persons are already implementing or planning to implement management strategies for improving and protecting water quality, but may need to take additional actions to meet the mercury TMDL allocations. 13.2.3 Identification of designated management agencies and responsible persons Identification of DMAs and responsible persons is required in the WQMP, as noted in OAR 340- 042-0040(4)(l)(G). The purpose of this element is to identify responsible persons and DMAs that are responsible for implementing the Willamette Basin Mercury TMDL. DMAs are federal, state and local Responsible persons include but are not limited to governmental agencies that have Designated Management Agencies. All responsible legal authority over an activity or persons, including DMAs, must implement the source contributing pollutants. DMAs TMDL. are identified as such by the Department of Environmental Quality in a TMDL. A responsible person is an entity, other than a federal, state, or local government, that is identified in a TMDL and has responsibility to meet assigned allocations and/or surrogate measures. Responsible persons may not have governmental (regulatory) authority to develop ordinances or other legal controls over activities. However, responsible persons identified in a WQMP may cause or contribute pollutant loading and have direct control over land or water management activities affecting mercury loading to rivers and streams. DMAs and responsible persons are responsible for implementing management strategies through existing regulatory

Oregon Department of Environmental Quality 72 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Figure 13-3. Number of Designated Management Agencies and Responsible Persons Identified in the Willamette Basin Mercury TMDL

and non-regulatory programs and activities and developing and revising sector-specific or source-specific implementation plans. DMAs and responsible persons are required to develop or revise TMDL implementation plans that describe the management measures they will take to achieve their load allocations (Section 13.3).

See Appendix E: List of designated management agencies and responsible persons for a complete list of DMAs and responsible persons named in the Willamette Basin Mercury TMDL. Appendix E is not intended to be an exhaustive list of every entity that bears responsibility for improving water quality in the Willamette Basin. All citizens that live, work and recreate in the Willamette Basin can take steps to reduce mercury and protect water quality. It will take broad participation to accelerate water quality improvements throughout the basin. 13.3 Management strategies This section of the plan describes management measures, as required in 340-042-0040(4)(l)(C), to reduce loadings of mercury to Willamette Basin waterbodies to meet TMDL load and wasteload allocations. It is organized by nonpoint and point source DMAs and responsible persons. For some of the DMAs, DEQ included a list of management measures as an implementation or “good practice” baseline. Generally, the list is not intended to be comprehensive or prescriptive and DMAs and responsible persons may propose alternative approaches or management strategies. Some of the management measures outlined in this section are already being implemented by DMAs, while others should be considered for future implementation. DEQ developed specific, required management practices related to cities and counties. Those requirements are found in Section 13.3.1.11 DEQ also developed specific requirements for impoundments described in Section 13.3.1.24.

Following the issuance of the 2006 Willamette Basin TMDL and WQMP, DEQ required individual DMAs to develop implementation plans that included specific management strategies and best management practices to meet load allocations for mercury. Reporting requirements

Oregon Department of Environmental Quality 73 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

for many of these DMAs included an annual progress report and a comprehensive assessment of activities every five years. Summaries and reports of implementation activities since the issuance of the 2006 TMDL are summarized below (Table 13-1).

All DMAs and responsible persons named in this TMDL will be required to either update or develop mercury reduction strategies and milestones as identified in Section 13.3.1. Note that riparian protection practices identified in the 2006 Willamette Basin that were related to temperature impairments can also help filter runoff and erosion of sediment containing mercury into waterbodies.. Together, these practices will provide a comprehensive approach to mercury pollution reduction. Existing information related to DMAs’ TMDL implementation efforts is available on DEQ’s websites below. Implementation plans and reports are also available on DMA websites.

Table 13-1. TMDL implementation reports and summaries

DMA TMDL Report Information available on DEQs website Oregon Biennial Agricultural Department of Water Quality Agriculture Management Area Plans https://www.oregon.gov/deq/wq/programs/wqstatustrends Oregon Biennial Agricultural Department of Water Quality Environmental Status and Trends Quality Analysis Oregon Willamette Basin Department of TMDL Five Year https://www.oregon.gov/deq/wq/tmdls/Pages/TMDLs- Environmental Review: DMA Implementation.aspx Quality Implementation 2008 - 2013 Oregon Oregon Nonpoint Department of Source Pollution https://www.oregon.gov/deq/wq/programs/Pages/Nonpoint.aspx Environmental Program Annual Quality Report Urban and TMDL annual Some DMAs provide a copy of their annual report and five year Rural DMAs progress report review report on their city or county website. These reports are also available from DEQ through a public records request. Urban and TMDL five year Rural DMAs review report Public records request information: https://www.oregon.gov/deq/Requesting-Public-Records

13.3.1 Management strategies for nonpoint sources and water protection programs As required in OAR 340-042-0040(4)(l)(E), the following section describes management strategies for nonpoint sources that will protect water quality. The section is arranged to include DMAs and responsible parties by state agencies, local governments, federal agencies and special districts.

Oregon Department of Environmental Quality 74 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

13.3.1.1 Oregon Department of Environmental Quality Nonpoint Source DEQ has the responsibility of overseeing and implementing Oregon’s Nonpoint Source Management Program Plan. A nonpoint source of pollution is any pollution entering a waterbody that does not come directly from a discrete conveyance. Nonpoint sources are typically not covered by NPDES permits. The goal of DEQ's Nonpoint Source Management Program is to reduce water pollution from nonpoint sources, in order to meet water quality standards. The nonpoint source program is implemented by coordinating with many local, state and federal agencies and organizations throughout Oregon. The program uses a combination of federal and state programs for implementing statewide, programmatic, and geographic priorities, objectives, and strategies to achieve short- and long-term goals. Program requirements include tracking and reporting on implementation actions and water quality outcomes from these activities in Oregon’s Nonpoint Source Annual Report submitted to EPA, which can be accessed on DEQ’s website: https://www.oregon.gov/deq/wq/programs/Pages/Nonpoint.aspx.

Oregon's Nonpoint Source Management Program is an important part of the state's water pollution control programs because for many pollutants, nonpoint sources of pollution are the major sources of pollution to a waterbody. A summary of DEQ programs that have the potential to reduce nonpoint source mercury loading in the Willamette Basin is provided Table 13-2.

Table 13-2. Summary of DEQ programs that have the potential to reduce mercury loading in the Willamette Basin.

DEQ NPS Program How it Protects/ Supports Water Quality Outlines and implements management goals, projects, and water Nonpoint Source TMDL quality monitoring for pollutant reductions that are needed in order Implementation Program meet Oregon’s water quality standards, including mercury and methylmercury. Protects human health and the environment by establishing Onsite Program requirements for the construction, alteration, repair, operation and maintenance of onsite wastewater treatment systems. Protects human health and the environment by identifying, Clean Up Program investigating, and remediating sites contaminated with hazardous substances, including mercury. The 319-grant program funds cooperating entities for activities that address nonpoint sources of pollution by emphasizing watershed protection and enhancement, watershed restoration, voluntary Nonpoint Source 319 Grant stewardship, and partnerships among watershed stakeholders, Program such as DEQ’s Pesticide Stewardship Partnership. This includes alignment with significant match funding provided through the Oregon Watershed Enhancement Board (OWEB)’s parallel granting programs. SRF loans finance a variety of nonpoint source water quality plans and projects. Eligible activities include integrated and stormwater Clean Water State Revolving management plans, establishing or restoring permanent riparian Fund buffers and floodplains and daylighting streams from underground pipes.

Oregon Department of Environmental Quality 75 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

13.3.1.2 DEQ Cleanup Program—Abandoned Mine Lands Sites The Cleanup program includes a number of subprograms, including Site Assessment. For a complete list of subprograms visit https://www.oregon.gov/deq/Hazards-and-Cleanup/env- cleanup/Pages/default.aspx. Site Assessment is responsible for screening abandoned mine lands sites to determine which sites may have significant impacts to the environment. Within the Willamette Basin there are 12 abandoned mine lands sites that were identified as significant sources of mercury, as shown in Table 9-2 of the Source Assessment Section in the TMDL. These sites represent legacy mines that were in operation prior to Oregon’s 1972 Oregon Mined Land Reclamation Act, and are now considered sources of “uncontrolled hazardous substances.” These sites are subject to statutes and rules administered by the Cleanup program (ORS 465; OAR 340.122).

Between 2000 and 2004, the Cleanup program collaborated with EPA, the federal Bureau of Land Management and the U.S. Forest Service to perform preliminary assessments of all abandoned mine lands sites in Oregon. Since that time, agency partners have completed site investigations, evaluations of potential cleanup levels and actions (feasibility studies), and the removal or treatment of contaminated materials. For up to date information, visit DEQ’s Environmental Cleanup Site Information database at https://www.deq.state.or.us/lq/ECSI/ecsiquery.asp.

DEQ also assists EPA in evaluating and implementing phased remediation of the former Black Butte mercury mine, which is listed on the National Priorities List as a federal Superfund site. See Section 13.3.1.16 for more information about remediation of this site as a leading factor in implementation of the legacy mining allocation.

13.3.1.3 DEQ Cleanup Program—Portland Harbor Superfund Source Control Portland Harbor is a heavily industrialized stretch of the Lower Willamette River north of downtown Portland, from Sauvie Island south to the Broadway Bridge. EPA listed Portland Harbor on the National Priorities List, known as Superfund, in December 2000 due primarily to contaminated sediment.

EPA, DEQ and other agencies, tribal governments, community groups and companies are working to investigate and clean up contamination in Portland Harbor. EPA is the lead agency responsible for investigating and cleaning up contaminated sediments in the river, while DEQ is the lead agency for investigating and cleaning up contamination on upland sites.

Although EPA and DEQ identified mercury as a contaminant of concern in this area, additional data and investigations to date show that mercury levels alone do not warrant active cleanup of particular sediment areas. However, upland remediation and planned in-water cleanup necessary for dioxins, pesticides, metals, polychlorinated biphenyls, and polycyclic aromatic hydrocarbons will also address some areas with mercury contamination. DEQ and EPA will be monitoring for mercury and relying on natural recovery to reduce concentrations in sediment and fish tissue. EPA established a cleanup level for mercury in fish tissue at 0.031 mg/kg. Additional information about Portland Harbor cleanup activities can be accessed on DEQ’s website: https://www.oregon.gov/deq/Hazards-and-Cleanup/CleanupSites/Pages/Portland- Harbor.aspx

Oregon Department of Environmental Quality 76 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

13.3.1.4 Oregon Department of Agriculture The responsibility of Oregon Department of Agriculture for regulating agricultural activities that impact water quality qualifies ODA as a DMA under OAR 340-042-080(3). The Agricultural Water Quality Management Act (ORS 568.900 to 933), and ORS 561.191, give ODA the responsibility to adopt and enforce rules that protect water quality on agricultural lands. The Agricultural Water Quality Management Act directs ODA to develop Agricultural Water Quality Management Area Plans as well as rules. Together, area rules and plans represent the two main pathways through which ODA implements TMDLs on non-federal agricultural lands in Oregon. DEQ will continue to work closely with ODA’s Agricultural Water Quality Program to ensure that ODA’s plans and rules are protective of water quality standards, including allocations and any surrogate measures contained in TMDLs. DEQ works with ODA as described under a 2012 Memorandum of Agreement.

Voluntary implementation through Agricultural Water Quality Management Area Plans ODA’s area plans identify local watershed conditions, water quality concerns associated with agriculture, and resources and strategies to address these concerns. There are a total of 38 Area Plans in Oregon, 10 of which specifically address watersheds within the Willamette Basin. These area plans include the Lower Willamette, Lower Columbia-Sandy, Clackamas, Middle Willamette, Molalla-Pudding-French Prairie- North Santiam, Tualatin, South Santiam, Southern Willamette, Upper Willamette- Upper Siuslaw, and Yamhill. Area plans are developed in consultation with Local Advisory Committees, which are made up of local farmers, and other watershed stakeholders.

ODA reviews each area plan on a biennial basis in consultation with local Soil and Water Conservation Districts, as well as the Local Advisory Committee. DEQ consults with ODA during the biennial review process to assess the water quality status and trends in the area in relation to allocations and any surrogate measures in an applicable TMDL. As part of the consultation process, DEQ provides a Status and Trends Report for each agricultural management area. These reports provide data and analysis of water quality status and trends in relation to water quality standards and TMDL allocations. ODA uses these reports to help identify implementation priorities at the catchment or watershed scale. Status and Trends Reports can be accessed from DEQ’s website.

After the biennial review process, the Local Advisory Committee submits progress reports to the Board of Agriculture and ODA Director. These reports will continue to include statistics on landowner engagement and types of management practices being employed. These reports will continue to be available to DEQ for review in assessing implementation progress.

Soil and Water Conservation Districts also continue to be key partners in implementing area plans. During the 2013-2015 biennium all Soil and Water Conservation Districts in Oregon started working in Focus Areas, which are geographic areas that are selected based on identified needs for agricultural water quality improvements. Soil and Water Conservation Districts contact agricultural landowners and offer voluntary assistance to improve streamside vegetation, streambank stability, and other concerns including livestock manure management and sediment reduction. These efforts are typically included in area plans and are evaluated as part the biennial review process.

Oregon Department of Environmental Quality 77 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Regulatory implementation through Agricultural Water Quality Management Area Rules The Agricultural Water Quality Management Act (ORS 568.900 – 568.933) describes how ODA may implement agricultural water quality management plans and rules. ORS 568.909(2) states that once ODA has designated the boundaries of a plan area, “…the department shall develop and carry out State agencies identified as designated a water quality management plan for management agencies must develop and implement the prevention and control of water measurable objectives related to reducing mercury pollution from agricultural activities and and mercury-related pollutants such as sediment. soil erosion. The department shall base the plan and rules adopted to implement the plan upon scientific information.” ODA has adopted rules that detail requirements on all agricultural lands. The rules describe the outcomes that must be achieved, which provides flexibility on how to achieve compliance with the outcomes.

Between 1998 and 2014, the Agricultural Water Quality Program primarily conducted compliance investigations based on written complaints received from the public and complaint referrals from other agencies. In 2014, ODA initiated Strategic Implementation Areas, which represent a proactive approach to identifying specific agricultural activities in a specific watershed that are violating ODA rules, as well as legacy conditions that are adversely affecting water quality, and identifying conservation actions that will help achieve water quality goals.

Strategic Implementation Area watersheds are designated by ODA after conferring with watershed partners including DEQ, and reviewing available water quality and other data. After establishing a Strategic Implementation Area, properties of concern within the Strategic Implementation Area are identified. After an initial assessment, ODA contacts landowners to offer assistance and determine compliance with local rules. For more information about Strategic Implementation Areas, visit https://www.oregon.gov/ODA. ODA will increase the number of SIAs they initiate every year beginning in 2020 and are actively working to obtain additional funding to grow the SIA program to include more areas.

ODA is the agency responsible for compliance investigations and enforcement of program rules, however Soil and Water Conservation Districts, Oregon Watershed Enhancement Board, US Department of Agriculture Farm Service Agency, US Department of Agriculture Natural Resources Conservation Service, watershed councils, and other partners also work to provide technical assistance and other resources to help landowners implement conservation activities.

In addition to the efforts described above, ODA also registers, administers and enforces water quality permits for Confined Animal Feeding Operations. ODA and DEQ jointly issue Water Pollution Control Facility state permits and NPDES federal permits for Confined Animal Feeding Operations. These permits do not allow discharges to waters of the state.

Measurable Objectives and WQMP Reporting Requirements For the purpose of this TMDL, ODA has identified minimizing bare ground as the strategy most likely to have the greatest impact on sediment and erosion, especially during wet winter months. In addition to minimizing bare ground, best management practices and conservation practices that limit livestock access to the riparian area, establish stream canopy, and help stabilize channel banks should be given the highest priority. Because stream crossings, road prism failures, and hydrologically-connected roads are known sources of sediment to waterbodies

Oregon Department of Environmental Quality 78 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

across land uses, DEQ expects to work with ODA to develop measurable objectives related to roads and a schedule for implementing these strategies following the issuance of the TMDL. Examples of such strategies include: inventorying hydrologically-connected roads and potentially unstable road prisms and at-risk stream crossings.

Management strategies that minimize the impact of agricultural activities on water quality are currently identified in area plans. Management strategies that specifically impact sediment and erosion are shown in Table 13-3.

Table 13-3. Example list of management strategies that are currently included in Agricultural Water Quality Management Area Plans to address sediment and erosion. Riparian Areas and Streams Resource Potential Benefits Potential Costs of Practice Concerns of Practice to Practice to Addressed Producer Producer Rotational grazing in May help establish Allows limited use of Requires intense riparian area; timed desirable riparian riparian area for management to insure when growth is vegetation and grazing, improves that grazing does not palatable to animals and address temperature wildlife habitat. prevent site capable when riparian area soils and bacteria TMDLs. vegetation from are not saturated. establishing. Livestock exclusion from Helps promote May lessen May require higher riparian area; desirable riparian streambank erosion weed control costs in establishing off-stream vegetation; promotes and loss of pastures; riparian areas than watering facilities. streambank integrity; less time involved in seasonal riparian helps filter nutrients managing livestock grazing. May require and sediment from grazing in riparian area, financial investment for runoff; may help improves wildlife livestock control and narrow channel and habitat. off-stream watering reduce erosion in facilities. channel and address temperature, mercury and bacteria TMDLs. Planting perennial Helps establish May lessen Costs of vegetation and vegetation in riparian perennial riparian streambank erosion weed control. May area. vegetation rapidly; and loss of pastures. If require financial promotes streambank livestock are excluded investment for riparian integrity; may help from riparian area, area fencing and off-stream narrow channel and may be eligible for watering facilities while reduce erosion in federal cost-share vegetation establishes. channel; provides programs. Some appropriate shade alternative perennial necessary to agricultural products moderate solar may be harvested from heating and address riparian areas. temperature, mercury and bacteria TMDLs.

Oregon Department of Environmental Quality 79 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Erosion, Sediment, and Mercury Control Resource Benefits to Practice Concerns Costs to Producer Producer Addressed Grazing management: Helps prevent May improve pasture Cost of installing graze pasture plants to sediment, nutrient, production; easy fencing, watering appropriate heights, mercury and bacteria access to water may facilities for rotational rotate animals between runoff into waters of increase livestock grazing system; time several pastures; the state. Helps production as well. May involved in moving provide access to water protect streamside improve livestock animals through in each pasture. areas. health because of pastures. better nutrition and parasite control. May improve composition of pasture plants and help prevent weed problems. Farm road construction: Helps prevent May help prevent water Cost of installation and construct fords sediment and mercury damage on farm roads. maintenance. appropriately, install runoff to waters of the water bars or rolling dips state. to divert runoff to roadside ditches. Plant appropriate Helps prevent May help prevent ditch Costs of establishing vegetation along sediment and mercury bank erosion and vegetation. drainage ditches; seed runoff into waters of slumping. ditches following the state. construction. Plant cover crops on Helps prevent May reduce weed Costs of establishing erosion-sensitive areas. sediment and mercury problems; prevents loss cover crops; cover runoff into waters of of applied nutrients. crops may compromise the state; helps filter primary crop. nutrients and slow runoff. Irrigate pasture or crops Helps prevent May reduce costs of Installation/ according to soil irrigation return flow irrigation; may help maintenance cost. moisture and plant water and associated crop or pasture Monitoring time. needs. nutrients, sediment, production. and mercury to waters of the state. Install/maintain Helps prevent nutrient Decreases muddiness Cost of installation. diversions or French and mercury runoff and shortens saturation drains to prevent into waters of the period in protected unwanted drainage into state. areas. barnyards and animal heavy use areas.

In addition to continued implementation of the strategies provided in Table 13-3, ODA will work with Local Advisory Committees, in consultation with DEQ, to identify specific measurable objectives and timelines such as percent reduction in bare ground during wet months, along with associated implementation timelines for implementing best management practices and conservation practices that address runoff, sediment and erosion. ODA will work with Local Advisory Committees to report on these metrics during the biennial review process.

Oregon Department of Environmental Quality 80 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

DEQ is requesting that ODA and Local Advisory Committees include specific metrics for identified areas of agricultural lands that can be tracked consistently across all agricultural water quality management plan areas in the Willamette Basin. DEQ recognizes that farming practices and cropping systems vary across and within these areas; however there are relevant strategies for reducing runoff, sediment and erosion that apply universally to almost all agricultural lands, e.g. reduce bare ground. This approach does not replace developing and tracking area-specific measurable objectives.

Measurable objectives and timelines should be coordinated with biennial reviews of area plans to the extent possible, however DEQ expects measurable objectives and timelines to be incorporated into all Willamette Basin area plans within 18 months of the issuance of this TMDL. ODA will also take part in the Willamette Basin five year review. For more information about five year reviews, see section 13.4.1.

13.3.1.5 Oregon Department of Forestry Under OAR 340-042-080(2), the Oregon Department of Forestry is the DMA for water quality protection from nonpoint source discharges or pollutants resulting from forest operations on non-federal forestlands within the state. The Forest Practices Act sets expectations for water quality outcomes and prescribes required best management practices. The Forest Practices Act has provisions for both criminal and civil penalties if forest operators do not comply with water protection regulations. ODF rules relevant to protection of water quality and erosion control are found in the Oregon Administrative Rules referenced in Table 13-4.

Table 13-4. ODF Rules Related to Water Quality and Erosion Control Forestry Practice Rule Reference Treatment of Slash OAR-629-615-0000 through 629-615-0300 Stewardship Agreements OAR 629-021-0100 through 629-021-1100 Forest Road Construction and Maintenance OAR-629-625-0000 through 629-625-0700 Harvesting OAR 629-630-0000 through 629-630-0800 Water protection rules OAR 629-635-0000 through 629-660-0060 Reforestation OAR 629-610-0000 through 629-610-0090 Afforestation OAR 629-611-0000 through 629-611-0020

In addition to assuring compliance with the Forest Practices Act, ODF also employs other efforts and funding, such as landowner voluntary measures conducted as part of the Oregon Plan for Salmon and Watersheds, to help support ODF’s role in implementing the TMDL. ODF also delivers technical assistance and cost share funding to family forest landowners that support goals for water quality protection. See Table 13-5 for examples of management strategies that resource managers on non-federal land implement to meet Forest Practices Act regulations to control erosion and runoff.

DEQ will work with ODF to identify specific actions necessary to reduce sedimentation from non-federal forest lands, including both voluntary and regulatory actions. For example, ODF’s February 2012 guide to voluntary actions to protect threatened and endangered fish is a good resource for non-federal forest landowners who wish to implement practices that go beyond the current Forest Practices Act and rules. For additional information about ODF, visit: http://www.oregon.gov/ODF.

Oregon Department of Environmental Quality 81 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Table 13-5. Example list of management strategies that ODF currently implements to address sediment and erosion. Forestry Practice Description Implement Forest Practices Act • Prescriptive and outcome-based rules for forest operations • Notification system (FERNS) • Forest operation inspections conducted by Stewardship Foresters • Compliance monitoring • Education and outreach on FPA topics Protection/enhancement of riparian • Stream and water body classification zone, wetlands, seeps, etc. with • Prescriptive rules on vegetation retention, ground buffers equipment, road building restrictions in riparian management areas • Promote implementation and reporting of Oregon Plan voluntary measures • Deliver incentive programs to restore/enhance aquatic/riparian habitat (CREP, etc.) Conduct pre-harvest planning • Stewardship Forester notification review, pre-operation inspections, and recommendations for any additional BMPs • Delivery of incentive programs to promote stewardship and planning Replace/restore roads/culverts • Prescriptive rules for road construction, maintenance and decommissioning • Identification and replacement/repair of culverts, ditches and other drainage elements of active and inactive roads that are not functioning properly or at risk of failure. • Promote implementation and reporting of Oregon Plan voluntary measures • Cease active road use during wet weather when roads have deep ruts or covered by a layer of mud that results in visible increases in stream turbidity (OAR 629-625-0700) Stabilize stream banks • Prescriptive rules for vegetation retention in riparian management areas • Rules to minimize, avoid, restore or prohibit ground equipment, road building in or near channels or channel modification Uplands management • Prescriptive rules for reforestation and harvesting • Rules to minimize soil disturbance and erosion and maintain productivity • Delivery of incentive programs to encourage forest health, minimize fire risk Inspection/enforcement • Civil Penalties • Forest operation inspections conducted by Stewardship Foresters BMP monitoring and evaluation • Adaptive management: effectiveness monitoring informs Board of Forestry who can revise prescriptive rules • Monitoring Strategy to prioritize and direct monitoring work Instream monitoring • Member of Water Quality Pesticide Management Team, Pesticide Stewardship Partnerships • Project-level instream water quality monitoring efforts to assess FPA effectiveness BMP implementation monitoring • Compliance audit study and reports

Oregon Department of Environmental Quality 82 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Forestry Practice Description Partnership for Forestry Education; • Delivery of technical assistance and cost share programs to education and outreach to operators family forest landowners and landowners • Agreement with Associated Oregon Loggers • Regional Forest Practices Committee • Committee for Family Forestlands • Partnership for Forest Education • Logging Conference session(s) • Annual Tree School events • Stewardship Forester delivery of individual landowner, operator technical assistance • Ad hoc training events: Operator breakfasts, Society of American Forester meetings, Watershed Council meetings, new rule training, etc.

The Memorandum of Understanding between ODF and DEQ describes a process to evaluate the sufficiency of current Forest Practices Act best management practices in meeting water quality standards and TMDLs on state and non-federally owned forestlands. Forest operators conducting operations in accordance with the Forest Practices Act are generally considered to be in compliance with water quality standards. Where it is shown that existing Forest Practice Act rules and voluntary measures are not sufficient to meet water quality standards, including TMDL load allocations, DEQ will request that ODF implement additional voluntary programs, revise statewide Forest Practices Act rules and/or adopt subbasin specific rules as necessary.

Measurable objectives, milestones, and WQMP reporting requirements In addition to continued implementation of the strategies provided in Table 13-5, and other voluntary efforts, DEQ and ODF will identify specific measurable objectives with milestones and associated implementation timelines that address runoff and erosion. Because stream crossings, road prism failures, and hydrologically-connected roads are known sources of sediment to waterbodies across land uses, DEQ expects to work with ODF to develop measurable objectives related to roads and a schedule for implementing these strategies following the issuance of the TMDL. Examples of such strategies include: inventorying hydrologically-connected roads and potentially unstable road prisms and at-risk stream crossings. Measurable objectives may also include an evaluation of hillslope erosion potential during tethered logging operations.

The measurable objectives and the metrics used for tracking measurable objectives will be submitted to DEQ in an implementation plan within 18 months of TMDL issuance.

Status of management strategies related to the Forest Practices Act erosion and runoff control requirements, progress on meeting milestones, and other ODF reporting, such as Forest Practices Compliance Audits will be included in subsequent Willamette Basin five year reviews. For more information about five year reviews, see Section 13.4. Reports or other documents used for ODF TMDL reporting should be made available on a publically accessible website.

Oregon Department of Environmental Quality 83 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

13.3.1.6 Oregon Department of State Lands Oregon Department of State Lands is named as a Designated Management Agency because DSL manages significant tracts of land and issues permits for earthwork below ordinary high water of waterways and in wetlands in the Willamette Basin. DSL’s authorities are noted in OAR 340-042-080(4).

DSL has both a regulatory and a proprietary role with regard to the land within the Willamette Basin. DSL issues two types of permits and authorizations related to its regulatory and proprietary roles: removal-fill permits for removal or fill activity in waterways and wetlands, and proprietary waterway authorizations for use of state-owned waterways.

In its regulatory role, DSL is responsible for administering Oregon’s Removal-Fill Law which was enacted in 1967 and includes the following responsibilities: • Protect, conserve and make best use of water resources • Protect public navigation, fishery and recreational areas • Ensure that activities of one landowner don’t adversely affect another landowner • Minimize flooding, improve water quality, and provide fish and wildlife habitat.

For many removal-fill permits, applicants also must obtain a corresponding permit from the U.S. Army Corps of Engineers under section 404 of the federal Clean Water Act. For these permits, DEQ issues water quality certifications under section 401 of the CWA. In its proprietary role, DSL owns certain state-owned parcels within the Willamette Basin, including: • Approximately 2,900 acres of land which includes both the surface and underlying mineral rights • Approximately 12,100 acres of mineral rights which occur on land on which the surface is owned by another entity (commonly termed “split estates”) • Submerged and submersible land underlying:

o The Willamette River from its confluence with the Columbia River at River Mile (RM) 0.0 to RM 187 at the confluence of the Coast and Middle Forks of the waterway; o The McKenzie River from its confluence with the Willamette River at RM 0.0 to RM 37 at Dutch Henry Rock; and o Tidally-influenced waters. As the manager of both upland parcels and mineral rights within the Willamette Basin, as well as submerged and submersible land underlying the Willamette River, DSL is responsible for authorizing uses placed on these holdings. Mercury occurs on all state-owned lands, including: • Upland parcels: primarily derived from local and distant sources by atmospheric deposition, and associated with possible underlying mineralization. • Submerged and submersible land: via atmospheric deposition and from runoff from upland and industrial discharges, and prior mining operations. • Mineral Rights: as an accessory constituent of, or used to process some mineral deposits.

Oregon Department of Environmental Quality 84 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Measurable Objectives, Milestones, and Water Quality Management Plan Reporting Requirements

DSL will continue to implement the management strategies identified in Table 13-6. These strategies help to ensure that all persons applying for, and holding authorizations to use, state- owned land are implementing best management practices that are protective of water quality.

In addition to the strategies identified in Table 13-6, DEQ encourages DSL to implement additional strategies, such as those shown in Table 13-7, to specifically reduce mercury and sediment movement. DEQ also encourages DSL to work with ODA and other watershed partners to conduct focused outreach and education that includes the water conveyance systems that are identified as responsible persons in this WQMP.

DSL is required to develop a TMDL implementation plan for the Willamette Basin for review and approval by DEQ within 18 months of the issuance of this TMDL. This plan must include specific measurable objective(s) and timelines for implementation and may include specific conditions that DSL and/or DEQ (through section 401conditions) utilize to avoid soil erosion and sedimentation. DSL will also take part in the Willamette Basin five year review. For more information about five year reviews, see Section 13.4.

Table 13-6. Management Strategies that Department of State Lands currently requires of people that apply for, and hold authorizations to use, state-owned land, which are protective of water quality in the Willamette Basin. Management Strategies

Dispose of all waste in a proper manner and do not allow debris, garbage or other refuse to accumulate. Do not cut, destroy or remove, or permit to be cut, destroyed or removed, any vegetation except with written permission. Conduct all operations in a manner which conserves fish and wildlife habitat, protects water quality, and does not contribute to soil erosion or the growth of noxious weeds. Maintain all buildings, docks, pilings, floats, gangways, similar structures, or other improvements (each an “Improvement”) in a good state or repair. Obtain any permits required by state or local authorities and comply with DEQ and Oregon State Marine Board requirements for sewage collection and waste water disposal for boats and floating structures. Do not use, store, or dispose of, or allow the use, storage, or disposal of any material that may pose a threat to human health or the environment except in strict compliance with applicable laws.

Oregon Department of Environmental Quality 85 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Table 13-7. Example list of management strategies to reduce sediment and erosion. Management Strategies

Maintain all structures, waste disposal and septic systems, and storm water runoff collection systems in good working condition. Condition or do not allow uses of submerged and submersible land that result in streambank erosion Encourage persons authorized to use state-owned land for grazing to prevent their animals from walking in or drinking directly from streams on state-owned property. In 2019, there are currently no rangeland class or agricultural class properties in the Willamette Basin. Not authorize any use of either upland or submerged and submersible land managed by the agency that involves the use of mercury or compounds containing mercury in amounts determined to be unacceptable based on comments received from the public review process of the application Not allow any use to occur on, or be made of state-owned submerged and submersible land that is determined to cause the release of an unacceptable amount of mercury from the sediments to the environment based on comments received from the public review process of the application Not allow any state-owned mineral deposit managed by DSL to be mined for mercury, or mercury to be used on state-owned land to process minerals Wherever possible, condition authorizations to limit or prevent stormwater runoff from, and resultant erosion of soil on state-owned land Clean up solid waste and other materials dumped illegally on state-owned land that may contain mercury, and attempt to identify the person(s) responsible for such activities for possible citation Employ interagency cross checks to confirm that a proposed use will not negatively impact a restoration site

Oregon Department of Environmental Quality 86 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

13.3.1.7 Oregon Parks and Recreation Department Under OAR 340-042-080(4), Oregon Parks and Recreation Department qualifies as a DMA due to responsibilities for managing several categories of lands owned by the state. Many of these areas remain undeveloped and while primarily managed for recreational uses, they also include lands managed for forestry and agriculture, such as livestock grazing. OPRD manages and operates over 130 individual parks, waysides and greenway properties, and more than 90 sites are leased to other entities for management. State Parks, State Natural Areas as well as upland areas are also managed by OPRD.

In 2017, OPRD released a 10-year Strategic Action Plan for restoration and stewardship of OPRD-managed sites in the Willamette Basin. The strategic plan, as well as a number of other programs and policies, integrate water quality implementation goals and objectives into existing management strategies, including: • Agricultural Use of Park Lands • Comprehensive Park Planning • Forest Management • Intergovernment Natural Resource Communications • Invasive Species Management on State Park Lands • Land Acquisition and Exchange • Maintenance and Operation of Water and Sewerage Systems • Natural Resource and Environmental Management Figure 13-4. Middle Fork Willamette River. Policy • Oregon Plan • Sewer and Water System Failures OPRD also administers a grant program and the State Scenic Waterways program, which support activities that are protective of water quality.

Measurable objectives, milestones, and WQMP reporting requirements OPRD’s TMDL implementation plan was recently updated in 2018 and includes multiple management strategies and actions that address mercury load reductions, including but not limited to those provided in Table 13-8. OPRD will continue to implement these and other management strategies in order to ensure that OPRD as well as all persons applying for, and holding authorizations to use state-owned land managed by OPRD are implementing best management practices that reduce runoff, sediment and erosion.

Oregon Department of Environmental Quality 87 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

In addition, OPRD will update their TMDL implementation plan to include specific measurable objectives, milestones and timelines for management strategies that address runoff and soil erosion within 18 months of the issuance of this TMDL.

OPRD will also take part in the Willamette Basin five year review. For more information about five year reviews, see Section 13.4.

Table 13-8. Example list of management strategies that OPRD currently implements to address sediment and erosion. Management Strategies

Continually monitor trail systems; repair or re-route trails to reduce runoff and erosion Continue to require permittees with Agricultural Leases to apply best management practices to prevent and reduce runoff and erosion, including retaining 50 foot no-till buffers along fish-bearing streams, and maintaining ground cover during wet, winter months Reduce number of drain tile systems in former agriculture fields to promote infiltration of stormwater Continue to meet or exceed all Forest Practices Act rules during forestry operations. Implement riparian restoration projects, which help to filter and reduce sediment delivery to streams Use on-site stormwater retention in new park designs to infiltrate stormwater Continue to provide education and outreach activities including promoting biking and walking to reduce air emissions

13.3.1.8 Oregon Department of Geology and Mineral Industries Under OAR 340-042-080(4), responsibility for regulation of aggregate mines, many of which are located in the flood plain of rivers, qualifies Department of Geology and Mineral Industries as a DMA. As with other state agencies that have been identified as DMAs, DOGAMI is required to submit an implementation plan specific to mercury reduction in the Willamette Basin, however, because DOGAMI conducts these activities throughout the state, DOGAMI may work with DEQ to develop a state-wide implementation plan to address other TMDL implementation responsibilities. Many of the elements required in an implementation plan will be met through DOGAMI’s oversite, as DEQ’s Agent of implementation of the NPDES 1200A general industrial stormwater permit. The 1200A permit covers aggregate and asphalt operations. Other elements required in an implementation plan are included in DOGAMI’s Best Management Practices for Reclaiming Surface Mines, which can be accessed on DOGAMI’s website: https://www.oregongeology.org/mlrr/overview.htm.

13.3.1.9 Oregon Department of Fish and Wildlife Per OAR 340-042-080(4), DEQ named Oregon Department of Fish and Wildlife DMA. ODFW manages three wildlife areas in the Willamette Basin, including EE Wilson Wildlife Area near Monmouth, Fern Ridge Wildlife Area near Eugene, and Sauvie Island Wildlife Area/ North Willamette Watershed Wildlife District near Portland. In addition to providing for wildlife habitat, these areas are also managed for recreational activities such as hunting, fishing, hiking, boating, wildlife observation, trapshooting and archery.

Oregon Department of Environmental Quality 88 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Measurable objectives, milestones, and WQMP reporting requirements ODFW will develop an implementation plan that will include management strategies and actions that address mercury load reductions, including but not limited to those provided in Table 13-9. ODFW will implement these and other management strategies in order to ensure that ODFW, as well as all persons applying for, and holding authorizations to use, ODFW owned land are implementing best management practices that reduce runoff, sediment and erosion.

In addition, ODFW’s implementation plan will include specific measurable objectives, milestones and timelines for management strategies that address runoff and soil erosion within 18 months of the issuance of this TMDL.

ODFW will also take part in the Willamette Basin five year review. For more information about five year reviews, see Section 13.4.

Table 13-9. Example list of management strategies that ODFW currently implements to address sediment and erosion. Management Strategies

Continually monitor trail systems; repair or re-route trails to reduce runoff and erosion Continue to require permittees with Agricultural Leases to apply best management practices to prevent and reduce runoff and erosion, including retaining 50 foot no-till buffers along fish-bearing streams, and maintaining ground cover during wet, winter months Reduce number of drain tile systems in former agriculture fields to promote infiltration of stormwater Continue to meet or exceed all Forest Practices Act rules during forestry operations. Implement riparian restoration projects, which help to filter and reduce sediment delivery to streams Use on-site stormwater retention in new park designs to infiltrate stormwater

13.3.1.10 Oregon State Marine Board Using authorities described in OAR 340-042-080(4), the Oregon State Marine Board administers boating safety educational programs, enforces marine law and maintains and improves boating facilities. OSMB establishes state-wide boating regulations and contracts with county sheriffs and the Oregon State Police to enforce marine laws. The board provides technical training to marine patrol officers and supplies their equipment. OSMB also provides grants and engineering services to local governments such as cities, counties, park districts and port districts, to develop and maintain accessible boating facilities and protect water quality. OSMB actively promotes safe and sustainable boating through several programs.

DEQ will coordinate with OSMB regarding implementation of the TMDL as it relates to boating practices. Boating activities potentially important to the implementation of the mercury TMDL include but are not limited to signage and education, establishment of boating regulations, practices for the removal of derelict structures that qualify under the Abandoned Vessel Program rules, and boating campaigns that encourage boaters to adopt clean and safe boating practices.

Oregon Department of Environmental Quality 89 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Measurable objectives, milestones, and WQMP reporting requirements OSMB will develop an implementation plan that will include management strategies and actions that address mercury load reductions. These management strategies will likely focus on boating practices.

In addition, the OSMB implementation plan will include specific measurable objectives, milestones and timelines for management strategies that address runoff and soil erosion related to boating practices within 18 months of the issuance of this TMDL.

OSMB will also take part in the Willamette Basin five year review. For more information about five year reviews, see section 13.4.

13.3.1.11 Cities Oregon cities and counties have the authority to regulate land use activities through local comprehensive plans and related development regulations. The Oregon land use planning system, which is administered by local governments with oversight through the Oregon Department of Land Conservation and Development, provides a unique framework for local jurisdictions to address water quality protection and enhancement. Every city and county is required to have a comprehensive plan and accompanying development ordinance to comply with state land use planning goals. While the comprehensive plan must serve to implement the statewide planning goals mandated by state law, cities and counties have a wide degree of local control over how resource protection is addressed in their community.

Many of the land use planning goals in OAR 660-015-0000 have a direct connection to water quality, particularly Goal 5 (natural resources, scenic, and historic areas and open spaces), Goal 6 (air, water, and land resources quality), and Goal 7 (areas subject to natural hazards). DEQ expects that the efforts of local jurisdictions to address Goals 5, 6, and 7 requirements, when incorporated into a TMDL implementation plan, will help a DMA meet the TMDL allocations. In addition, existing city and county efforts to protect and enhance riparian vegetation along streams will reduce sediment movement by stabilizing streams banks and providing natural filtering of runoff containing sediment.

Mercury in Urban Stormwater

TMDL modelling shows that in urban areas, the majority of mercury reaches waterbodies through atmospheric deposition and through runoff of mercury from soils and hard surfaces. TMDL modelling has limitations in separating natural and background nonpoint sources of mercury and sediment from agriculture, forestry, non-MS4 permitted urban areas and other nonpoint sources. While agriculture and forestry land uses can be dominant in some rural areas, commercial and industrial activities, as well as large housing developments, are prevalent in many rural areas of the basin. Therefore, DEQ expects that city and county DMAs will largely focus on activities and strategies to reduce runoff and erosion into urban and rural streams and into stormwater conveyance systems.

During the first implementation phase of the 2006 Willamette Basin Mercury TMDL, DEQ required some MS4 Phase I communities to collect mercury stormwater data. DEQ analyzed total mercury data from seven MS4 Phase I communities (see Table 13-10).

Oregon Department of Environmental Quality 90 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

The TMDL water column target to meet a fish tissue methylmercury criterion of 0.040 mg/kg is 0.14 ng/L. The median value of total mercury in stormwater from the MS4 Phase I communities was 4.62 ng/L. Based on the analyzed data, DEQ concluded that urban stormwater has environmentally significant concentrations of mercury contributing to mercury loads in portions of the Willamette Basin, even though the sector’s overall load to the basin is small. Therefore, to reduce mercury from urban runoff to the maximum extent practicable, DEQ developed point source wasteload allocations for NPDES MS4 permit holders, and nonpoint source load allocations for non-MS4 permitted urban DMAs—both set at a 75% reduction of total mercury.

Table 13-10. Stormwater Summary Statistics Median 25th % 75th % Analyte Sample Size Range (ng/L) (ng/L) (ng/L) (ng/L) Total Hg 655 0.25 - 120 4.62 2.94 8.31 Source: Tetra Tech, 2019a

Six Minimum Measures for Cities to Reduce Stormwater Impacts to Waterbodies

EPA established six stormwater control measures, previously only required for MS4 Phase I entities, as part of its final EPA MS4 Phase II stormwater regulations (January 9, 1998 63 FR 1536 -1643). These regulations acknowledge that uncontrolled urban stormwater presents a significant water quality problem. Implementing the six measures provides a consistent set of stormwater control measures for all regulated MS4 permit holders to decrease delivery of stormwater pollutants from urban runoff.

Requirements in DEQ’s MS4 Phase II general permit that became effective in March 2019, generally mirrors the six EPA control measures. For this TMDL, DEQ will also defer to these six stormwater control measures to control unpermitted urban runoff from cities with a population of 5,000 or greater. DEQ is requiring implementation of these stormwater measures to achieve needed nonpoint DEQ is applying six stormwater control measures to source reductions in mercury and many urban areas. sediment. At the same time, DEQ recognizes that implementing these requirements will not only reduce sediment, but also have benefits in reducing other urban pollutants associated with stormwater.

The six stormwater control measures described below in Table 13-11 are generally less prescriptive than the requirements contained in DEQ’s MS4 Phase II general permit. Applying these measures to all urban areas, including those not previously regulated by a permit or TMDL requirements, fills an implementation gap to ensure mercury and sediment in stormwater discharges are controlled throughout the Willamette Basin to the maximum extent practicable.

Oregon Department of Environmental Quality 91 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Table 13-11. Minimum requirements for cities. In addition to requirements in section 13.3.2.2., these requirements apply to cities with MS4 permits (outside of the MS4 permit coverage area) and to cities without a MS4 permit Stormwater Requirements Measure 1. Pollution DMAs must properly operate and maintain its facilities, using prudent Prevention and pollution prevention and good housekeeping to reduce the discharge of Good mercury-related pollutants, such as sediment, through the stormwater Housekeeping for conveyance system to waters of the state. Municipal Operations DMAs must ensure that DMA-owned or operated facilities with industrial activity identified in DEQ’s 1200-Z Industrial Stormwater General Permit have coverage under this permit. The DMA must also conduct its municipal operation and maintenance activities in a manner that reduces the discharge of pollutants to protect water quality.

DMAs must maintain records for activities to meet the requirements of the Pollution Prevention and Good Housekeeping for Municipal Operations program requirements and include a descriptive summary of their activities in the TMDL Annual Report. 2. Public Education DMAs must conduct an ongoing education and outreach program to inform and Outreach the public about the impacts of stormwater discharges on waterbodies and the steps that they can take to reduce mercury-related pollutants in stormwater runoff. The education and outreach program must address stormwater issues of significance within the DMA’s community.

DMAs must track implementation of the public education and outreach requirements. In each corresponding TMDL Annual Report, the DMA must assess their progress toward implementation of the program, including a qualitative evaluation of at least one education and outreach activity corresponding to the reporting timeframe for the associated TMDL Annual Report. The evaluation should be used to inform future stormwater education and outreach efforts to most effectively convey the educational material to the target audiences. 3. Public DMAs must implement a public involvement and participation program that Involvement and provides opportunities for the public to effectively participate in the Participation development of stormwater control measures. The DMA must comply with their public notice requirements when implementing a public involvement participation process, including maintaining and promoting at least one publicly accessible website with information on the city’s stormwater control implementation, contact information and educational materials. 4. Illicit Discharge DMAs must implement and enforce a program to detect and eliminate illicit Detection and discharges into the stormwater conveyance system. An illicit discharge is Elimination any discharge to a stormwater conveyance system that is not composed entirely of stormwater. The DMA must develop and maintain a current map of their stormwater conveyance system. The stormwater conveyance system map and digital inventory must include the location of outfalls and an outfall inventory, conveyance system and stormwater control locations. The DMA must make maps and inventories available to DEQ upon request. When in digital format, the DMA must fully describe mapping standards in the TMDL implementation plan or other city planning document.

The IDDE program must prohibit non-stormwater discharges into the stormwater conveyance system through enforcement of an ordinance or other legal mechanism, including appropriate enforcement procedures and actions to ensure compliance. The ordinance or other regulatory mechanism

Oregon Department of Environmental Quality 92 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Stormwater Requirements Measure must also define the range of illicit discharges it covers, including those discharges that are conditionally allowed, such as groundwater and lawn watering discharges. The IDDE program must also maintain a procedure or system to document all complaints or reports of illicit discharges into and from the stormwater conveyance system.

The DMA must track implementation of the IDDE program requirements. In each TMDL Annual Report, the DMA must assess their progress towards implementation of the program. 5. Construction Site DMAs must refer project sites to DEQ, or the appropriate DEQ agent, to Runoff Control obtain NPDES 1200-C Construction Stormwater Permit coverage for construction projects that disturb one or more acres (or that disturb less than one acre, if it is part of a “common plan of development or sale” disturbing one or more acres).

In addition, DMAs must require construction site operators to complete and implement an Erosion and Sediment Control Plan for construction project sites in its jurisdictional area that result in a minimum land disturbance of 21,780 square feet (one half of an acre) or more, and are not already covered by a 1200-C permit.

Through ordinance or other regulatory mechanism, to the extent allowable under state law, the DMA must require erosion controls, sediment controls, and waste materials management controls to be used and maintained at all qualifying construction projects (as described above) from initial clearing through final stabilization to reduce pollutants in stormwater discharges to the stormwater conveyance system from construction sites.

The DMA must develop, implement and maintain a written escalating enforcement and response procedure for all qualifying construction sites. The procedure must address repeat violations through progressively stricter response, as needed, to achieve compliance.

The DMA must track implementation of its construction site runoff program required activities. In each TMDL annual report, the DMA must assess their progress toward implementing its construction site runoff program’s control measures. 6. Post-Construction DMAs must develop, implement, and enforce a program to reduce Site Runoff for discharges of pollutants and control post-construction stormwater runoff from New Development new development and redevelopment project sites in its jurisdictional area. and Example of such programs and program elements are provided in Appendix Redevelopment D.

Through ordinance or other regulatory mechanism, the DMA must require the following for project sites discharging stormwater to the storm water conveyance system that create or replace 10,890 square feet (one quarter of an acre) or more of new impervious surface area: (A) The use of stormwater controls at all qualifying sites. (B) A site-specific stormwater management approach that targets natural surface or predevelopment hydrological function through the installation and long-term operation and maintenance of stormwater controls.

Oregon Department of Environmental Quality 93 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Stormwater Requirements Measure (C) Long-term operation and maintenance of stormwater controls at project sites that are under the ownership of a private entity. The DMA must target natural surface or predevelopment hydrologic function to retain rainfall on-site and minimize the offsite discharge of precipitation utilizing stormwater controls that infiltrate and evapotranspirate stormwater. For projects that are unable to fully retain rainfall/runoff from impervious surfaces on-site, the remainder of the rainfall/runoff from impervious surfaces must be treated prior to discharge with structural stormwater controls. These stormwater structural controls should be designed to remove, at a minimum, 80 percent of the total suspended solids.

The DMA must maintain records for activities to meet the requirements of the post-construction site runoff program requirements and include a descriptive summary of their activities in the TMDL Annual Report.

Nonpoint source requirements for city and special districts with MS4 permits Cities and certain special districts with MS4 Phase I or Phase II permits will meet point source TMDL requirements for reducing mercury by implementing these permits. See Section 13.3.2.2 for these specific requirements.

Cities and special districts that currently have MS4 Phase I or Phase II stormwater permits within the Willamette Basin are listed in Table 9-5. Note that some special districts have the authority for implementing MS4 permit requirements for cities.

As DMAs for nonpoint sources of mercury, MS4 permit holders must also implement the six stormwater control measures, as described in Table 13-11in their jurisdictional areas outside of the urbanized area covered by their permit. If city jurisdictional boundaries include land uses under the authority of other DMAs, such as Oregon Department of Transportation, Oregon Department of Agriculture, or Oregon Department of Forestry, then those DMAs are responsible for stormwater discharge from these properties. Also, note that some cities have existing pollution control authorities, such as regulations and permits associated with the Underground Injection Program to protect discharges to groundwater. The stormwater control measures contained in this section do not replace these regulations. The DMA should contact DEQ if they have questions about the applicability of required stormwater control measures in areas where other related regulations exist.

While the six minimum stormwater measures in Table 13-11 are less rigorous than MS4 permit requirements, MS4 permit holders may choose to implement requirements under their permit within and outside the urbanized area of their permit for implementation consistency. This approach would either meet or exceed the requirements in Table 13-11. Similarly, some MS4 permit holders apply all or a portion of their stormwater control requirements outside their permit coverage area, so may already meet certain control measures. As another example, it is likely that all of the MS4 Phase II cities have urbanized areas that extend beyond their MS4 boundaries, so the MS4 permit would cover all stormwater requirements. In these cases, the DMA must include this important information in their TMDL implementation plan update and clearly articulate how the TMDL requirements are being met. The overall goal is to ensure that urban areas in MS4 jurisdictions are implementing, at a minimum, the six stormwater control measures identified in Table 13-11.

Oregon Department of Environmental Quality 94 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

In addition, TMDL implementation plans for cities and special districts that have MS4 permits must demonstrate how nonpoint source load allocations for mercury will be met by including management strategies to reduce runoff and erosion that discharge directly to waterbodies. This requirement is to ensure that cities are implementing a comprehensive approach to reducing sediment and mercury not only to the MS4 conveyance system, but also from sources directly adjacent to waterbodies. While DEQ maintains authority to address discharges to waters of the state, cities and counties also have authority to control land uses within their jurisdiction that have the potential to negatively impact water quality.

Reporting on point source and nonpoint source TMDL implementation may be streamlined into a single submission, which will be reviewed by both DEQ stormwater and TMDL program staff. See Measurable objectives, milestones, and WQMP reporting requirements section for more information about updating TMDL implementation plans and reporting.

Nonpoint source requirements for cities without MS4 permits The requirements for cities that have stormwater discharges and are not required to have MS4 permit coverage are discussed below. If a city is later identified by DEQ as needing coverage under an MS4 permit, the MS4 permit requirements would supersede the requirements below within the permit coverage area.

DEQ does not have stormwater mercury data from discharges occurring in cities that are not regulated by a MS4 Phase I permit. In the absence of data, DEQ cannot quantitatively determine the amount of mercury in stormwater discharges from these smaller cities. However, analyses show that mercury contained in stormwater is primarily a function of runoff and erosion from impervious areas, rather than from specific sources in large urban areas, and could contribute to a water quality impairment. In addition, the percent of impervious cover in the Willamette Basin continues to increase in almost all jurisdictions, as seen in the number of municipal building permits and DEQ 1200-C construction permits. For this reason, DEQ is requiring cities with a population of 5,000 people or greater to implement stormwater control measures.

Note that the 2006 Willamette Basin TMDL already required cities with populations greater than 10,000 people to implement the six stormwater control measures to reduce mercury and bacteria loads from urban areas. While these DMAs were required to implement the six stormwater control measures, this TMDL provides more specificity in what these measures must entail.

The stormwater requirements described in Table 13-11 will apply throughout the city boundary for areas not under the jurisdiction of another federal or state agency such as Oregon Department of Transportation, Oregon Department of Agriculture, Oregon Department of Forestry, Bureau of Land Management, or U.S. Forest Service. Additional details about implementing the six stormwater control measures based on population status are provided below.

Cities with populations 5,000 people or greater The following cities meet a population criterion of 5,000 people or greater (according to Portland State University July 1, 2018 certified dataset): • Greater than 10,000: (1) Canby, (2) Cottage Grove, (3) Dallas, (4) Lebanon, (5) McMinnville, (6) Newberg, (7) St. Helens, (8) Woodburn, (9) Sandy, (10) Silverton

Oregon Department of Environmental Quality 95 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

• 5,000 – 10,000: (1) Creswell, (2) Independence, (3) Junction City, (4) Molalla, (5) Monmouth, (6) Scappoose, (7) Sheridan, (8) Stayton and (9) Sweet Home. These communities will need to either develop a new TMDL implementation plan, or update their existing TMDL implementation plan to fully incorporate the stormwater measures for mercury and sediment reduction described in Table 13-11. Alternatively, if DMAs can demonstrate to DEQ that their existing TMDL implementation plan meets requirements stated in this WQMP, no revisions are necessary. Cities named above must implement the six stormwater control measures according to the schedule in Table 13-11.

Cities with populations less than 5,000 people City DMAs with a population of less than 5,000 people must evaluate the six minimum stormwater control measures listed in Table 13-11 and identify the strategies and actions that they can implement to reduce mercury and sediment, including sources of runoff, sediment and erosion. Upon request, DMAs must provide information to DEQ regarding their specific limitations to implementing all or some of the six stormwater controls.

Under certain circumstances, such as when population growth exceeds 5,000 people or DEQ determines it is necessary to meet load allocations for mercury, DEQ may in the future require cities with a population less than 5,000 people to implement all or a subset of the six stormwater control measures.

If identified, these communities will need to either develop a new TMDL implementation plan, or update their existing TMDL implementation plan to include strategies that address stormwater runoff and erosion.

13.3.1.12 Counties Counties pose a unique situation in terms of land use and pollutant sources because of a mix of densely populated urban areas with sparsely populated rural areas. For these reasons, WQMP requirements for counties are tailored to focus on specific strategies and best management practices that are relevant to reducing mercury and sediment loading within county jurisdiction. Counties will apply the strategies and best management practices found in Table 13-12 to all county-owned lands, properties, facilities and roads, as applicable.

If a county is already implementing any of the management programs specified in Table 13-12, the county must include this important information in their TMDL implementation plan and clearly articulate how TMDL requirements are being met. Other land uses in the county that are under the authority of other DMAs for TMDL implementation, such as cities, Oregon Department of Transportation, Oregon Department of Agriculture, Oregon Department of Forestry, or U.S. Forest Service and others, will be managed according to each agency’s TMDL requirements.

Counties are included with other nonpoint source sectors, such as agriculture and forestry, to meet the aggregated 88% total mercury reduction load allocation. The 75% total mercury load reduction required for city stormwater discharges only applies within permit boundaries for those counties with MS4 permits.

Oregon Department of Environmental Quality 96 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Table 13-12. Minimum management program requirements for counties. These requirements apply to county MS4 permit holders outside of the MS4 permit coverage area, and throughout the county jurisdiction for counties without MS4 permits. Management Requirements Program 1. Pollution Counties must properly operate and maintain lands, properties, and facilities Prevention and using prudent pollution prevention and good housekeeping measures, and Good through appropriate staff training reduce the discharge of mercury-related Housekeeping for pollutants to waterbodies. County Operations Counties must maintain records for meeting these requirements and include a descriptive summary of their activities in the TMDL annual report. 2. Public Education Counties must conduct public education and outreach to reduce mercury and Outreach and mercury-related pollutants, such as sediment, on county lands and properties, as applicable. Such activities should include outreach to property owners adjacent to county roads and ditches. In addition, public outreach must include efforts to encourage and facilitate reporting of sediment related issues or concerns from the public. Public outreach should be tailored to meet the needs and diversity of the county population (e.g. signs, social media, website presence, etc.).

Counties must track implementation of the public education and outreach requirements and describe all activities in the TMDL annual report. 3. Enforcement of Counties must reduce conveyance of mercury and mercury-related Prohibited pollutants to waterbodies from county lands and properties, and have Pollutants capability of enforcing on other entities that contribute mercury-related pollutants, such as sediment, to county property and assets.

DEQ recognizes that county ordinances already in place or that must be adopted will likely be more comprehensive and prohibit discharges of other pollutants, rather than only those pollutants associated with mercury.

The program must also maintain a procedure or system to document all complaints or reports of mercury and mercury-related pollutant discharges to county lands and properties and to water bodies from county lands and properties.

In each TMDL annual report, counties must track implementation of their enforcement program and describe all activities. 4. Construction Site To minimize mercury and control potential sediment runoff from construction Runoff Control sites, counties must incorporate erosion control requirements into county building and grading permit applications. Permit language must require erosion, sediment and waste material management controls to be used and maintained at construction sites from initial clearing through final stabilization. Counties may prioritize where these building and grading permit requirements are applied, for example where increased development is occurring, according to county zoning regulations, or where large subdivisions or large-scale dense development is allowed.

Through an ordinance or other regulatory mechanism, counties must be able to pursue enforcement and technical assistance, as appropriate, at construction sites where pollutants could discharge to waters of the state, either directly to stream or through a conveyance system.

In each TMDL annual report, the county must track implementation of its construction site runoff control program and describe all activities.

Oregon Department of Environmental Quality 97 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

County roads and associated ditches have a high potential to convey sediment and runoff to waterbodies if not managed properly, therefore DEQ expects to see specific management measures related to roads in county TMDL implementation plans. Examples of road-related issues and other important strategies and best management practices to reduce erosion and runoff are included in Table 13-13 below.

Table 13-13. Example List of County Management Strategies to Reduce Sediment/Mercury Management Strategies

Roads • Identify and prioritize county roads and ditches that contribute sediment and runoff to waterbodies. Best practices could include planting/retaining vegetation in ditches and reducing use of pesticides when appropriate to site conditions. Special attention should be focused on the following situations: unimproved/gravel roads in higher traffic areas; roads where traffic consists of heavy machinery use; near quarries or other activities that can exacerbate dust and track-out concerns. • Develop a stormwater management plan to ensure that pre-construction and post-construction stormwater runoff from roads, highways and bridges is treated prior to discharge to a waterbody. • Prioritize and replace undersize or improperly designed culverts that can increase flow velocities and cause erosion of sediment around the culvert and ditch. • Appropriate road construction design to reduce erosion at stream crossings and accommodate flood events • Develop and implement an operation and maintenance program with a schedule of regular and long term inspection and maintenance ensuring the proper operation and effectiveness of both structural and source controls, e.g. stormwater system maintenance and/or road maintenance actions that prevent erosion of road surfaces

Riparian Buffers • Retain or plant adequate riparian buffers along waterbodies on county properties, such a parks to provide natural filtering of sediment. Percent effective shade targets to meet the 2006 Willamette Basin TMDL for temperature are available in the 2006 TMDL document. Meeting shade targets will help provide shade for reducing heat impacts, as well as buffers to filter runoff. Counties that were identified as DMAs in the 2006 TMDL should already be conducting activities in support of this goal. • Develop an enforceable ordinance that establishes a minimum buffer along streams, wetlands, lakes, and other waterbodies.

Onsite Stormwater Management • Counties should strive to reduce the percent of new impervious surfaces by prioritizing onsite stormwater infiltration on county-owned properties for existing properties, new development and redevelopment. • Counties should encourage other developers to implement low-impact design standards on large development sites.

Oregon Department of Environmental Quality 98 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Nonpoint source requirements for counties with MS4 permits Counties that have Phase I or Phase II MS4 permits will meet point source TMDL requirements for reducing mercury by implementing its permit following issuance of the TMDL and subsequent DEQ approval of its revised permit. See section 13.3.2.2 for these specific requirements.

Counties with MS4 permits must implement management programs Cities and Counties must make their TMDL identified in Table 13-12 on county lands, implementation plans available to the public properties and facilities outside the urbanized area of the permit, and report on the status of these management programs in TMDL annual reports.

In addition to management programs in Table 13-12, TMDL implementation plans for county DMAs that have MS4 permits must contain mercury reduction strategies, such as those listed in Table 13-13. In addition, TMDL implementation plans must demonstrate how nonpoint source load allocations for mercury will be met by including management strategies to reduce runoff and erosion that discharge directly to waterbodies. This requirement is to ensure that counties are implementing a comprehensive approach to reducing sediment and mercury not only to the MS4 conveyance system within their permit coverage area, but also from sources directly adjacent to waterbodies. While DEQ maintains authority to address discharges to waters of the state, counties can also have authority to control land uses within their jurisdiction that have the potential to negatively impact water quality.

Counties that currently have MS4 Phase I or Phase II stormwater permits within the Willamette Basin are listed in Table 9-5.

Nonpoint source requirements for counties without MS4 permits Counties that do not currently have MS4 permits must implement management programs identified in Table 13-11 on county lands, properties and facilities within the entire jurisdiction of the county, and report on the status of these management programs in TMDL annual reports. County lands falling outside the Willamette Basin watershed do not apply. The only counties in the Willamette Basin that do not have MS4 permits at this time are Yamhill and Columbia counties. If one of these counties is later identified by DEQ as needing coverage under an MS4 permit, the MS4 permit requirements would supersede the requirements in Table 13-12 within the permit coverage area.

TMDL implementation plans for counties must contain mercury reduction strategies, such as those listed in Table 13-13.

Implementation Schedule for Cities and Counties

Since the issuance of the 2006 Willamette Basin TMDL, some city and county DMAs have already been implementing BMPs to reduce mercury and sediment inputs to the watershed, including, but not limited to: • Conducting outreach and education about best management practices for the management of dental wastes and recycling of fluorescent lighting • Requiring sediment and erosion control plans of new and re-development projects

Oregon Department of Environmental Quality 99 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

• Requiring or encouraging the use of low impact development to reduce the volume and rate of stormwater discharged to streams • Reducing emissions by purchasing more fuel-efficient vehicles for municipal fleets • Enforcing and/or encouraging conservation and enhancement of riparian buffers, which trap sediment and prevent stream bank erosion • Performing regular street sweeping and catch basin cleaning Because of the greater reductions in mercury necessary for this TMDL, cities and counties will need to conduct additional implementation actions. At the same time, DEQ also recognizes the financial challenges that cities and counties face in implementing the Willamette Basin TMDLs. For this reason, DEQ is proposing to allow cities and counties additional time to fully implement mercury control measures in TMDL implementation plans.

See Table 13-14 for the implementation schedule for cities. See Table 13-15 for the implementation schedule for counties. If a city or county cannot meet these deadlines, please discuss your circumstances with the DEQ basin coordinator for your watershed.

Table 13-14. Stormwater Control Measures Implementation Schedule for Cities. Implementation Deadlines from TMDL Issuance Date Stormwater City Population Control Measures Less than 5,000 5,000 to 10,000 Greater than 10,000 1. Pollution Prevention and As determined by DEQ Good based on information 3 years 18 months Housekeeping provided by DMA for Municipal Operations 2. Public Education As determined by DEQ and Outreach based on information 3 years 18 months provided by DMA 3. Public As determined by DEQ Involvement and based on information 3 years 18 months Participation provided by DMA 4. Illicit Discharge As determined by DEQ Detection and based on information 4.5 years 3 years Elimination provided by DMA 5. Construction Site As determined by DEQ Runoff Control based on information 9.5 years 4.5 years provided by DMA 6. Post- Construction Site As determined by DEQ Runoff for New based on information 9.5 years 4.5 years Development and provided by DMA Redevelopment

Oregon Department of Environmental Quality 100 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Table 13-15. Management Program Implementation Schedule for Counties. Implementation Deadlines from TMDL Management Program Issuance Date 1. Pollution Prevention and Good 18 months Housekeeping for Municipal Operations 2. Public Education and Outreach 18 months 3. Enforcement of Prohibited Pollutants 3 years 4. Construction Site Runoff Control 4.5 years

Measurable objectives, milestones, and WQMP reporting requirements Cities and counties identified in Appendix E: List of designated management agencies and responsible persons must update or develop new TMDL implementation plans within 18 months following the issuance of the TMDL. DEQ may approve an alternate deadline, such as the due date associated with a DMA’s TMDL annual report to align with existing reporting deadlines. Alternatively, revisions will not be necessary if DMAs can demonstrate to DEQ that existing TMDL implementation plans meet requirements stated in this section.

TMDL implementation plans must contain milestones to meet specific requirements contained in Table 13-11 for cities with populations greater than 5,000 or Table 13-12 for counties. In addition, DMAs will develop measurable objectives associated with those requirements in order to track progress over time. Although there are no specific deadlines for cities with less than 5,000 people to implement stormwater control measures in Table 13-11. DEQ expects these DMAs to evaluate the six minimum stormwater control measures and identify the strategies and actions in TMDL implementation plans that they can implement to reduce mercury and sediment. Cities with less than 5,000 people must also provide information to DEQ about specific limitations to implementing strategies that the city does not include in its implementation plan.

Cities and counties that have a publically accessible website must post their implementation plan to that website. Cities and counties that do not have a publically accessible website must work with DEQ to make their plans publically accessible, such as providing copies by request at city hall or other offices where the public interacts with city staff, or through newsletters.

Cities or counties may continue to partner with other jurisdictions, as appropriate, to meet any requirements. If jurisdictions pursue these partnerships, DEQ encourages cities and counties to have formal agreements in place to ensure requirements are met given their legal status as a DMA. In addition, these details must be documented in the TMDL implementation plan.

Cities and counties will also take part in the Willamette Basin five-year review. For more information about five year reviews, see Section 13.4.

Appendix D contains a list of stormwater management resources to help cities develop TMDL implementation plans to address stormwater measures, including resources to assist cities in funding and developing post-construction stormwater ordinances and manuals. In addition, a number of cities in the Willamette Basin have had similar stormwater management requirements in place since the mid-1990’s based on their status as a MS4 permit holder and could be a wealth of information and potential watershed partners for other communities.

Oregon Department of Environmental Quality 101 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

13.3.1.13 Bureau of Land Management The federal Bureau of Land Management is responsible for management and regulation of Federal agencies identified as designated certain forest and range lands owned management agencies must develop and implement by the federal government. In measurable objectives related to reducing mercury western Oregon these are primarily and mercury-related pollutants such as sediment. forestlands. As a DMA in this TMDL, the BLM is required to develop and implement TMDL strategies and actions that address sediment, erosion and runoff.

The DEQ and BLM have a Memorandum of Understanding signed in 2017, which ensures water quality standards, TMDLs, and drinking water rules and regulations are met. The MOU also specifies that the BLM will implement site-specific best management practices as specified in management objectives, management direction, design features, and mitigation developed in Resource Management Plans and amendments, project-level plans, and Water Quality Restoration Plans to meet applicable water quality standards. The Resource Management Plans for western Oregon, which include the Willamette Basin, were recently updated in 2015. Water Quality Restoration Plans are the BLM’s implementation plan to meet TMDL requirements. Water Quality Restoration Plans exist for the following areas: Clackamas, Lower Willamette, Mid-Coast, Middle Willamette, Molalla, North, Santiam, Sandy, South Santiam, Tualatin, Upper Willamette, and Yamhill.

The MOU requires monitoring to ensure that practices are properly designed and applied, to determine the effectiveness of practices in meeting water quality standards, and to provide for adjustment of best management practices when it is found that water quality standards are not being protected.

Activities on BLM lands that contribute to sediment include transportation system management, recreation and forest management. Table 13-16 contains several examples of sediment, erosion and runoff control best management practices that address activities that occur on BLM lands. BLM will also continue to work with DEQ and EPA to assess and remediate, as warranted, the remaining abandoned mine lands on BLM properties. For more information about abandoned mine lands, see Section 9.2.3. The TMDL assigned legacy metal mines a 95% total mercury reduction.

The BLM incorporates water quality management as part of project design. Additionally, BLM employs best management practices that are relevant to the action in order to meet water quality standards and TMDL load allocations.

Best management practices are monitored for effectiveness following implementation. Appendix J of BLM’s Resource Management Plan provides a list of typical best management practices that the BLM uses to manage water quality. The BLM also designs site-specific best management practices to address specific issues and conditions that have the potential to affect water quality. The BLM will evaluate the effects of their management at the scale of the Willamette Basin.

Oregon Department of Environmental Quality 102 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Table 13-16. Example list of management strategies that BLM currently implements to reduce sediment and erosion. Best Management Practices

Design stream crossings to minimize diversion potential in the event that the crossing is blocked by debris during storm events. This protection could include hardening crossings, armoring fills, dipping grades, oversizing culverts, hardening inlets and outlets, and lowering the fill height. Disconnect road runoff to the stream channel by out sloping the road approach. Suspend ground-disturbing activity if forecasted rain will saturate soils to the extent that there is potential for movement of sediment from the road to wetlands, floodplains, and waters of the State. Road closure and decommissioning: After tilling the road surface, pull back unstable road fill and end- haul or contour to the natural slopes. Place residual slash on severely burned areas, where there is potential for sediment delivery into waterbodies, floodplains and wetlands. Emergency stabilization or rehabilitation BMPs related to wildfire Water bar spacing requirements by percent gradient and erosion class Implement erosion control measures at recreation sites to stabilize exposed soils where water flows or sediment, may reach waterbodies.

Locate new Off Highway Vehicle trails on stable locations (for example, ridge tops, benches, and gentle-to-moderate side slopes). Minimize trail construction on steep slopes where runoff could channel to a waterbody. Use erosion-reduction practices, such as seeding, mulching, silt fences, and woody debris placement, to limit erosion and transport of sediment to streams from quarries. Establish the Riparian Reserve along all intermittent and perennial streams, lakes, natural ponds, wetland, springs, seeps, impoundments and unstable areas that are adjacent to stream channels and are likely to deliver material to the stream in the event of a failure. The Riparian Reserve is a BLM land use allocation with specific management requirements to improve or maintain aquatic habitat and function. Work with DEQ and EPA to assess and remediate, as warranted, the remaining abandoned mine lands on BLM properties

Measurable objectives, milestones, and WQMP reporting requirements BLM will continue to implement their best management practices program. In addition, BLM will also identify specific measurable objectives with milestones and associated implementation timelines for implementing best management practices that address runoff and erosion. Because stream crossings, road prism failures, and hydrologically-connected roads are known sources of sediment to waterbodies across land uses, DEQ expects to work with BLM to develop measurable objectives related to roads and a schedule for implementing these strategies following the issuance of the TMDL. Examples of such strategies include: inventorying hydrologically-connected roads and potentially unstable road prisms and at-risk stream crossings. Measurable objectives may also include an evaluation of hillslope erosion potential during tethered logging operations.

A rationale, which provides context for the measurable objectives and the metrics used for tracking measurable objectives, will be submitted to DEQ within 18 months of TMDL issuance. The measurable objectives and milestones will be included in revised Water Quality Restoration

Oregon Department of Environmental Quality 103 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Plans based on either sixth field level watersheds (HUC12) or combined into one Water Quality Restoration Plan for the entire Willamette Basin. Water Quality Restoration Plan(s) must be made available on a publicly accessible website.

BLM will also take part in the Willamette Basin five year review. For more information about five year reviews, see Section 13.4.

13.3.1.14 U.S. Forest Service The United States Forest Service (within the US Department of Agriculture) is the federal agency tasked with the management and care of the National Forests and Grasslands. As a DMA in this TMDL, the USFS is required to develop and implement TMDL strategies and actions that address erosion and runoff.

A DEQ and USFS Memorandum of Understanding signed in 2014, identifies Water Quality Restoration Plans as the implementation planning document to meet USFS TMDL implementation plan requirements. The USFS submits these Water Quality Restoration Plans to DEQ for review and approval. The memorandum specifies that USFS will provide an annual status to DEQ on Water Quality Restoration Plans, including a five-year report on implementing each WQRP. The most recent publication date of the Willamette Basin Water Quality Restoration Plan is 2008.

The USFS relies on the following mechanisms to support TMDL implementation: • Aquatic Conservation Strategy in the Northwest Forest Plan • National Core BMP Technical and Monitoring guides. There is a summary of a two-year effort to demonstrate and document best management practices performance. The National BMP Program provides a nationally consistent, systematic, and objective approach to best management practices monitoring on USFS lands. • The 2005 Travel Management Rule (36 CFR 212.5) directed all National Forests to identify the minimum road system needed for safe and efficient travel and for administration, utilization, and protection of National Forest System lands. The rule requires each National Forest to:

o Identify the minimum road system needed for safe and efficient travel and for administration, utilization, and protection of national forest lands; o Identify the roads on lands under Forest Service jurisdiction that are no longer needed to meet forest resource management objectives; o Under separate actions, decommission or consider for other uses those roads identified as unneeded.

The Mt. Hood, Willamette and Umpqua National Forests completed Travel Analysis Plans for their respective road systems by September 30, 2015. These high-level plans provided a starting point for right sizing road systems, balancing public use, administrative use and resource protection. All subsequent planning on National Forest lands within the Willamette Basin tiers to these Travel Analysis Plans to inform and prioritize road maintenance, reconstruction, storage and decommission.

Activities on USFS lands that contribute to sediment include transportation system management, recreation and forest management. Table 13-16 contains several examples of sediment, erosion and runoff control best management practices that address activities that

Oregon Department of Environmental Quality 104 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 occur on USFS lands. USFS will also continue to work with DEQ and EPA to assess and remediate, as warranted, the remaining abandoned mine lands on USFS properties. For more information about abandoned mine lands, see Section 9.2.3. The TMDL assigned legacy metal mines a 95% total mercury reduction.

Table 13-17. Example list of management strategies that USFS currently implements to reduce sediment and erosion. Best Management Practices

Roads • Design or reconstruct stream crossings to minimize diversion potential in the event that the crossing is blocked by debris during storm events. This protection could include hardening crossings, armoring fills, dipping grades, oversizing culverts and lowering the fill height. • Disconnect road runoff to the stream channel by either out sloping or adding additional drainage features to the road. • Road closure and decommissioning: depending on aquatic risk, treatment activities could range from water barring and berm closure, to removal of all fills/culverts to complete obliteration and re-contour.

Timber Harvest • To prevent sediment delivery to streams, prescribe adequate no-harvest buffers on both perennial and intermittent streams within treatment areas. • Suspend ground-based harvest activities during saturated soil conditions where there is potential for sediment delivery into waterbodies, floodplains and wetlands. • Dependent on road condition, suspend timber haul to prevent sediment delivery to waterbodies, floodplains and wetlands during wet weather.

Erosion Control Measures during Construction • Require a dewatering and erosion control plan for construction activities such as culvert replacement and aquatic restoration projects to prevent sedimentation to waterbodies, floodplains and wetlands to the greatest extent practicable.

Wildfire • Where there is potential for sediment delivery into waterbodies, floodplains and wetlands, obliterate (de-compact and re-contour) all direct and indirect dozer and hand lines constructed for emergency suppression after fire is controlled.

Abandoned Mine Lands • Work with DEQ and EPA to assess and remediate, as warranted, the remaining abandoned mine lands on USFS properties Measurable objectives, milestones, and WQMP reporting requirements In addition to continued implementation of the strategies provided in Table 13-17, the USFS will identify specific measurable objectives with milestones and an associated implementation timeline for implementing best management practices that address runoff and erosion. Because stream crossings, road prism failures, and hydrologically-connected roads are known sources of sediment to waterbodies across land uses, DEQ expects to work with USFS to develop measurable objectives related to roads and a schedule for implementing these strategies following the issuance of the TMDL. Examples of such strategies include: inventorying hydrologically-connected roads and potentially unstable road prisms and at-risk stream crossings. Measurable objectives may also include an evaluation of hillslope erosion potential during tethered logging operations.

Oregon Department of Environmental Quality 105 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

A rationale, which provides context for the measurable objectives and the metrics used for tracking measurable objectives, will be submitted to DEQ within 18 months of TMDL issuance. The measurable objectives and milestones will be included in revised Water Quality Restoration Plans based on either sixth field level watersheds (HUC12) or combined into one Water Quality Restoration Plan for the entire Willamette Basin. Water Quality Restoration Plans must be made available on a publicly accessible websites.

The USFS will also take part in the Willamette Basin five year review. For more information about five year reviews, see Section 13.4.

13.3.1.15 U.S. Fish and Wildlife Service The U.S. Fish and Wildlife Service is an agency that manages fish, wildlife and natural habitats. In the Willamette Basin, the USFWS manages four wildlife refuges, including WL Finley National Wildlife Refuge near Corvallis, Ankeny Wildlife preserve near Ankeny Wildlife Refuge near Jefferson, Baskett Slough Wildlife Refuge near Dallas, and Tualatin River National Wildlife Refuge near Wilsonville. In addition to providing wildlife habitat, these areas are also managed for recreational activities including hunting, wildlife observation and hiking.

Measurable objectives, milestones, and WQMP reporting requirements The USFWS will update their current implementation plan to include management strategies and actions that address mercury load reductions, including but not limited to those provided in Table 13-18. USFWS will implement these and other management strategies in order to ensure that USFWS, as well as all persons applying for, and holding authorizations to use, USFWS owned land are implementing best management practices that reduce runoff, sediment and erosion.

In addition, the USFWS implementation plan will include specific measurable objectives, milestones and timelines for management strategies that address runoff and soil erosion within 18 months of the issuance of this TMDL.

The USFWS will also take part in the Willamette Basin five year review. For more information about five year reviews, see Section 13.4.

Table 13-18. Example list of management strategies that USFWS currently implements to reduce sediment and erosion. Management Strategies Continually monitor trail systems; repair or re-route trails to reduce runoff and erosion. Continue to require permittees with Agricultural Leases to apply best management practices to prevent and reduce runoff and erosion, including retaining 50 foot no-till buffers along fish-bearing streams, and maintaining ground cover during wet, winter months. Reduce number of drain tile systems in former agriculture fields to promote infiltration of stormwater. Monitor and assess how water is managed on the refuges through ditches, pumps, weirs, lakes, etc.

Continue to meet or exceed all Forest Practices Act rules during forestry operations. Implement riparian restoration projects, which help to filter and reduce sediment delivery to streams. Use on-site stormwater retention in new park designs to infiltrate stormwater.

Oregon Department of Environmental Quality 106 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

13.3.1.16 U.S. Environmental Protection Agency Among other responsibilities, the U.S. Environmental Protection Agency regulates remediation of federal Superfund sites on the National Priorities List under the Comprehensive Environmental Response, Compensation and Liability Act. Within the Willamette Basin, EPA manages remedial action of the following Superfund sites that have the potential to reduce mercury entering waters of the basin: Black Butte Mine; Portland Harbor; and Gould Inc. The Black Butte Mine reductions fall under the Legacy Mine source category with 95% reduction needed. Portland Harbor and Gould fall under the general nonpoint source category with 88% reductions needed.

EPA is also responsible for national regulation of air emissions and as the primary scientific advisor to the U.S. Government on negotiating international treaties on mercury air emissions and deposition reductions. EPA regulates coal-fired power plants under section 112(b) of the Clean Air Act as sources of hazardous air pollutants, including mercury. In June 2019, EPA eliminated the agency’s Clean Power Plan and replaced it with the Affordable Clean Energy Rule that gives states more flexibility in requiring upgrades for coal-fired power plants in order to achieve greenhouse gas emissions reductions. While the Clean Power Plan required emission reductions to meet targets, Affordable Clean Energy Rule required only that coal-fired power plants demonstrate improved efficiency in fuel use. EPA’s analysis (in Table 6, 83 Federal Register 44784) showed that reduction in greenhouse gas and other pollutant emissions, such as nitrogen oxides and sulfur dioxide, would be much less under Affordable Clean Energy Rule than under Clean Power Plan. Another analysis concluded that Affordable Clean Energy Rule could lead to worse air quality conditions than repealing and not replacing the Clean Power Plan (Keyes, et al., 2018). For example, increased greenhouse gas and other pollutant emissions could result because more efficiently operating coal-fired power plants would be likely to operate more often and for a longer period of time. EPA also completed a national-scale risk assessment in 2011 based on coal- and oil-fired power plant emission contributions to mercury deposition (U.S. Environmental Protection Agency, 2011).

In the TMDL, the air deposition source sector is assigned an allocation of 11% reductions needed. Any EPA efforts to reduce national or global reductions in mercury would help contribute to the 11 percent reduction.

DEQ will work with EPA staff following the issuance of the TMDL to discuss how EPA will report to DEQ on progress in implementing remediation actions for these CERCLA sites and air emissions and deposition regulation and treaty negotiations that will all contribute to reductions of mercury within the Willamette Basin.

Special Districts 13.3.1.17 Metro (Portland Metropolitan Government) Metro is the regional government for the Portland metropolitan area. Metro manages the solid waste program, regional parks and natural areas system, coordinates growth in the metro area, and oversees large facilities, such as the Oregon Zoo, Oregon Convention Center and the Portland Expo Center.

Metro is currently a DMA for a number of Willamette Basin TMDLs. Metro’s activities include proposing bond measures to acquire natural areas. Parks and natural area levies allow for natural area restoration, such as tree and shrub planting, removal of invasive vegetation, and

Oregon Department of Environmental Quality 107 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

reconnecting rivers to their floodplains. Metro follows local MS4 permit requirements in construction and post construction for any new or redeveloped Metro projects.

Measurable objectives, milestones, and WQMP reporting requirements As a DMA for the mercury TMDL, DEQ will work with Metro following the issuance of the TMDL to focus on stormwater control activities that will reduce erosion and runoff of stormwater from Metro properties. In addition, Metro will identify specific measurable objectives with milestones and associated implementation timeline for implementing best management practices that address runoff and erosion. An updated implementation plan will be due 18 months following the issuance of the TMDL. Metro must post their implementation plan on a publicly accessible website.

13.3.1.18 Port of Portland The Port of Portland is a regional government with jurisdiction in Multnomah, Washington and Clackamas counties. Port of Portland property in the Lower Willamette Basin includes the Portland International and Hillsboro Airports, four marine terminals (Terminals 2, 4, 5 and 6), and the Swan Island, Rivergate, Portland International Center, and Cascade Station business and industrial parks. The Port also owns a number of undeveloped properties within the basin that include open space, mitigation areas, and industrial parcels for future development. Some of these properties are occupied by tenants, which have lease agreements with the Port.

The Port of Portland’s MS4 permit can serve as the implementation plan for the mercury TMDL for the MS4 permit applicable service area. In addition, the Port of Portland will also implement, or continue to implement, management strategies to reduce runoff and erosion from Port of Portland properties that could discharge mercury in stormwater directly to waterbodies in the Willamette Basin, as well as discharges through MS4-permitted conveyances. The Port of Portland must update its TMDL implementation plan to ensure that management measures to reduce erosion and runoff directly to waterbodies are included in their suite of pollutant reduction programs. Nonpoint source requirements also related to MS4 permit holders are found in Section 13.3.1.11. In addition, the Port of Portland must post its nonpoint source implementation plan to address areas not covered by their MS4 permit applicable service area on a publicly accessible website. Other NPDES permits held by the Port of Portland will be implemented according to requirements set forth in Section 13.3.2.2.

13.3.1.19 Clean Water Services Clean Water Services, which is a water resources management utility, manages four wastewater treatment plants and implements a single MS4 stormwater permit for its 12 co- implementers within the urban portions of Washington County.

CWS’s MS4 permit can serve as the implementation plan for the mercury TMDL for the MS4 permit applicable service area. In addition, CWS will also implement, or continue to implement, management strategies to reduce erosion and runoff within its stormwater service area that could discharge mercury in stormwater directly to waterbodies, in addition to discharges through MS4-permitted conveyances. CWS must update its TMDL Implementation plan to ensure that management measures to reduce erosion and runoff directly to waterbodies are included in their suite of pollutant reduction programs. Nonpoint source requirements also related to MS4 permit holders are found in Section 13.3.1.11. In addition, CWS must post its nonpoint source

Oregon Department of Environmental Quality 108 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Implementation plan on a publicly accessible website. Other NPDES permits held by CWS will be implemented according to requirements set forth in Section 13.3.2.2.

13.3.1.20 Tualatin Hills Park and Recreation District Tualatin Hills Park and Recreation District is responsible for managing over 2,000 acres of land in Washington County. THPRD is a special park and recreation service district funded primarily by property taxes and program fees. Its service area spans the City of Beaverton and many unincorporated areas of eastern Washington County. The district has 27 miles of streams and three lakes within its boundaries.

Measurable objectives, milestones, and WQMP reporting requirements THPRD will develop an Implementation plan that will include multiple management strategies and actions that address mercury load reductions, including but not limited to those provided in Table 13-19. THPRD will implement these and other management strategies in order to ensure that THPRD, as well as all persons applying for, and holding authorizations to use, THPRD owned land are implementing best management practices that reduce runoff, sediment and erosion.

In addition, THPRD will update their TMDL implementation plan to include specific measurable objectives, milestones and timelines for management strategies that address runoff and erosion within 18 months of the issuance of this TMDL.

Table 13-19. Example list of management strategies that THPRD currently implements to reduce sediment and erosion. Management Strategies

Continually monitor trail systems; repair or re-route trails to reduce runoff and erosion. Continue to meet or exceed all Forest Practices Act rules during forestry operations. Implement riparian restoration projects, which help to filter and reduce sediment delivery to streams. Use on-site stormwater retention in new park designs to infiltrate stormwater. Continue to provide education and outreach activities including promoting biking and walking to reduce air emissions.

13.3.1.21 Oak Lodge Water Services District Oak Lodge Water Services District provides drinking water, wastewater, and watershed protection services in Oak Grove, Jennings Lodge, and portions of Milwaukie and Gladstone.

OLWSD’s MS4 permit can serve as the implementation plan for the mercury TMDL for the MS4 permit applicable service area. In addition, OLWSD will also implement, or continue to implement, management strategies to reduce erosion and runoff from OLWSD properties that could discharge mercury in stormwater directly to waterbodies, in addition to discharges through MS4-permitted conveyances. OLWSD must update its TMDL implementation plan to ensure that management measures to reduce erosion and runoff directly to waterbodies are included in their suite of pollutant reduction programs. Nonpoint source requirements also related to MS4 permit holders are found in Section 13.3.1.11. In addition, OLWSD must post its nonpoint

Oregon Department of Environmental Quality 109 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

source implementation plan on a publicly accessible website. Other NPDES permits held by OLWSD will be implemented according to requirements in Section 13.3.2.2.

13.3.1.22 Water Environment Services Clackamas County Water Environment Services (WES) provides wastewater resource recovery services for more than 190,000 customers in northwest Clackamas County. WES manages and operates five wastewater treatment facilities and owns and maintains more than 300 miles of sanitary sewer pipes, 16 pump stations and three interties. WES also implements programs to protect streams from polluted stormwater runoff, restores habitat for fish and other wildlife, provides regulation and review of land development plans, and engages with the community and partners to protect watershed health. Every year, the WES Riverhealth Stewardship Program provides grants to organizations dedicated to improving the health of local watersheds.

Much of WES' Surface Water Management (SWM) service area is regulated by a Phase I MS4 Permit, but significant portions, such as the rural lands in the Johnson Creek watershed, are not in the area regulated by the MS4 permit and they are not in the WES service area. Other portions of the SWM service area are served by drywells and other types of shallow stormwater injection devices, which are regulated by a Stormwater Water Pollution Control Facilities permit issued by DEQ. Approximately 80,000 people live in the WES SWM service area. Nearly all of these people live in the area which is east of the City of Milwaukie, which includes Clackamas (97015) and Happy Valley, but approximately 2,000 people reside in the portion of the WES SWM service area which is located in the Tualatin River watershed. Rivers and streams in the WES SWM service area include Johnson Creek, Kellogg Creek, and the Tualatin and Clackamas Rivers.

WES’s MS4 permit can serve as the implementation plan for the mercury TMDL for the MS4 permit applicable service area. In addition, WES will also implement, or continue to implement, management strategies to reduce erosion and runoff within its stormwater service area that could discharge mercury in stormwater directly to waterbodies, in addition to discharges through MS4-permitted conveyances. WES must update its TMDL Implementation plan to ensure that management measures to reduce erosion and runoff directly to waterbodies are included in their suite of pollutant reduction programs. Nonpoint source requirements also related to MS4 permit holders are found in Section 13.3.1.11. In addition, WES must post its nonpoint source Implementation plan on a publicly accessible website. Other NPDES permits held by WES will be implemented according to requirements set forth in Section 13.3.2.2.

13.3.1.23 Water Delivery and Conveyance Systems Irrigation districts, drainage districts, and other water delivery and conveyance systems influence the quantity and timing of sediment delivery to downstream river reaches. Return flows can enter waters of the state through ditches and pipes. Consequently, owners and operators of these systems are included as responsible persons in this WQMP because maintenance and management of these systems can impact sediment transport and erosion. Such systems are responsible only for sedimentation resulting from conveyance systems, not from upland agricultural activities such as raising livestock, cultivation, and irrigation. These activities are covered by the Department of Agriculture as part of its responsibilities as a DMA (see Section 13.3.1.4)

Irrigated agriculture is the largest consumptive surface water use in the Willamette Basin, and the volume of water consumed is predicted to increase over the next 50 years. A USGS study

Oregon Department of Environmental Quality 110 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

found that more than 75 percent of water use in the Willamette Basin was derived from surface flow, and the largest single use was for irrigated agriculture. Growth in irrigation water rights leveled off in the 1990’s (Jaeger, Plantinga, Langpap, Bigelow, & Moore, 2017), however the US Army Corps of Engineers recently projected irrigated acres on lands already in agricultural production to increase by more than 70,000 acres between 2020- 2070 within their study area (US Army Corps of Engineers, 2017). While irrigated agriculture continues to be an important and potentially growing demand, there remains a need to characterize the location and extent of irrigation systems in the basin, as well as the management practices used to maintain and operate these systems. Drainage districts and systems exist primarily to manage stormwater drainage and flooding. Many of these districts were originally formed to help protect the land from flooding so that farming could occur year round. Presently, drainage districts that are registered with the state as special districts often have a tax base that comprise rural tracts of land, as well as commercial and residential properties and parks. Levees, pump stations, ditches, sloughs, streams and culverts are important components of a drainage system and must be continually maintained in order to protect the environment, property and safety.

Water conveyance systems, including those that are managed for irrigation and drainage, are currently regulated by multiple state and federal agencies, including Oregon Water Resources Department, DSL, USACE, and DEQ’s 401 water quality certification program. For most waters, a DSL permit is required if a project will involve 50 cubic yards of fill and/ or removal within the ordinary high water line of a stream; this requirement also applies to some ditches. Projects that require a DSL removal-fill permit may also require a Clean Water Act Section 404 permit from the USACE. For these projects, a joint application form can be submitted to both agencies. Existing regulatory programs relevant to these activities are summarized in Table 13-20.

Implementing the requirements and conditions of these permits and Water Quality Certifications include best management practices that meet the TMDL requirements. For projects and activities that are exempt or not permitted by the agencies and programs shown in Table 13-16, owners and operators of water conveyance systems must implement similar best management practices to reduce sediment and erosion, in order to meet the TMDL requirements.

Table 13-20. Existing state and federal agencies and programs that regulate water conveyance systems Regulatory permit or Agency Program certification U.S. Army Corp of Engineers Willamette Valley Project Water Service Contracts and U.S. Bureau of Reclamation DEQ 401 Program Water Quality Certification Department of State Lands Waterways and Wetlands Removal/ Fill Permit Oregon Water Resources Permits for withdrawal, Water Rights Department storage, and use

Appendix E: List of designated management agencies and responsible persons lists the water conveyance entities that DEQ has identified as responsible persons. Operation and maintenance of any hydro-modification system that discharges return flows to waters of the state has the potential to impact the timing and quantity of sediment delivery to streams, thus there remains a need to better characterize the geographic location and current operation and maintenance activities related to water conveyance entities in the Willamette Basin. This information will help DEQ and system owners and operators gain a better understanding of their potential impact on reducing sediment and erosion.

Oregon Department of Environmental Quality 111 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

There may be additional water conveyance systems in the Willamette Basin that are not included in Appendix B due to limited availability of information about existing systems. However, all systems that have return flows or the potential to discharge to waters of the state should implement management measures to reduce sediment and erosion. Measurable objectives, milestones, and WQMP reporting requirements DEQ developed proposed milestones and timelines for working with owners and operators of water conveyance systems (Table 13-21). DEQ will collaborate with watershed partners including ODA and Oregon Water Resources Congress to conduct outreach and education to water conveyance entities over the next two years. DEQ will also work individually with owners and operators of water conveyance systems to gather information and better characterize their potential to discharge or have return flows to the Willamette Basin river network and determine what management and reporting strategies are relevant to their specific operations and maintenance activities.

DEQ expects Water Conveyance entities identified in Appendix E: List of designated Owners and operators of water conveyance management agencies and responsible systems must implement best management persons to work with DEQ as outlined in Table practices related to operation and maintenance of 13-21. Examples of the types of management their system. strategies that responsible persons will be required to implement are shown in Table 13-22.

Table 13-21. Milestones and timelines for DEQ to work with water conveyance entities to plan and carry out implementation of the 2019 Willamette Basin Mercury TMDL Strategy Action Milestone Estimated Timeline Conduct outreach and DEQ will worked with Individually contact Completed: Initial education to water Oregon Department of Water Conveyance contact completed by conveyance systems in Agriculture, and Entities identified in June 30, 2019. the Willamette Basin, reached out to Oregon Appendix E of the specifically those Water Resources Willamette Basin Hg identified in Appendix E Congress and other TMDL WQMP using of the 2019 Willamette watershed partners to available contact Basin Mercury TMDL provide informational information. WQMP. and educational Provide at least one in- Informational meeting opportunities relevant person informational will occur in 2019 to the Willamette Basin meeting during the Hg TMDL. public comment period Work directly with DEQ will work with Complete at least one Meeting to occur Water Conveyance water conveyance in-person meeting after between Dec. 1, 2019 Entities to better entities to characterize the public comment and April 30, 2020. identify and and document water period. characterize water conveyance systems conveyance systems for purpose of identified in Appendix E identifying relevant of the 2019 Willamette management Basin Mercury TMDL strategies, and WQMP. implementation tracking and reporting requirements.

Oregon Department of Environmental Quality 112 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Strategy Action Milestone Estimated Timeline Work directly with Finalize implementation Schedule Implementation Water Conveyance strategies, objectives, implementation planning and Entities to develop timelines and reporting planning and development meetings implementation requirements. development meetings. to occur between May strategies, objectives, 1, 2020 and Sept. 30, and timelines, and 2021. reporting requirements. Water Conveyance All information to be Entities will submit submitted according to DEQ- requested schedule identified information that is during one-on-one and/ necessary to develop or aggregate implementation, implementation tracking and reporting planning and strategies and development meetings requirements. (see above). DEQ will finalize DEQ will finalize implementation, implementation, tracking and reporting tracking and reporting requirements. requirements by Dec. 31, 2021.

Table 13-22. Examples of Management Strategies that may be required of water conveyance entities named as responsible persons in the 2019 Willamette Basin Mercury TMDL WQMP Water Quality Protection Management Strategies for Water Conveyance Entities

List of turbidity/sediment control best management practices for watercourse maintenance activities. Maintain a list of construction or ditch maintenance activities that require state and/ or federal permits or ODFW approval. Use streambank and/ or canal stabilization practices, including structural and non-structural best management practices. Manage upland conveyance system infrastructure, for example, roads, pumps, etc. to prevent soil erosion, and sediment delivery to waterbodies. Conduct education and outreach to water users and upland agricultural and urban land owners that discharge to system. Monitor and evaluate best management practices and strategies. Flow and drainage management to reduce erosion, and sediment delivery to streams. Maintain a schedule for operation and maintenance activities. Maintain a current map of system, including canals, ditches, pumps, weirs, etc.

Oregon Department of Environmental Quality 113 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

13.3.1.24 Reservoir management Impoundments create conditions where mercury methylation rates are higher than flowing stream segments. Higher rates of MeHg production can result in reservoir fish having higher MeHg concentrations. There is also potential for release of methylmercury from impoundments to lower stream segments.

According to the Oregon Department of Water Resources dam inventory, there are 414 dams in the Willamette Basin that can store at least 9.2 acre-ft. Included in the inventory are dams defined by OAR 690-020-0022(8): “Dam” means hydraulic structure built above the natural ground line that is used to impound water. Dams include all appurtenant structures, and together are sometimes referred to as “the works”. Dams include wastewater lagoons and other hydraulic structures that store water, attenuate floods, and divert water into canals.

Collectively, Willamette Basin dams can store over 2.7 million acre-ft. Many of the dams are located in areas under the authority of various DMAs or responsible persons. Appendix E: List of designated management agencies and responsible persons shows dams, owners, responsible persons, and DMAs for the 124 dams storing at least 100 acre-ft, which is the smallest capacity of the dams owned by the four largest owners. All DMAs and responsible persons operating reservoirs must be aware of factors contributing to increased reservoir methylation rates, including water level fluctuations, thermal stratification, and nutrient loading to reservoirs. DMAs and responsible persons must also be familiar with the operations or conditions resulting in dam releases or discharges to surface water.

The U.S. Army Corps of Engineers, Portland General Electric, U.S. Bureau of Reclamation and Eugene Water and Electric Board are the four largest owners and operators of reservoirs in the Willamette Basin, based on maximum storage volumes. Reservoir implementation requirements pertaining to for these four DMAs/responsible persons are specified below in the section titled “Measurable objectives, milestones, and WQMP reporting requirements.”

Oregon Department of Environmental Quality 114 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Figure 13-5. Map of Reservoirs Belonging to the Four Largest Owners in the Willamette Basin

Oregon Department of Environmental Quality 115 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Table 13-23. Example of Best Management Practices for Reservoirs Best Management Practices Oxidant addition to reservoir bottom waters Hypolimnetic oxygenation systems In-reservoir sediment removal or encapsulation1 Artificial circulation Reduction of average water level fluctuations Vegetation management Sediment amendment 1This might be valuable in areas where the historic sediment in the reservoir has significantly higher mercury concentrations than contemporary sediment deposition

Measurable objectives, milestones, and WQMP reporting requirements USACE, PGE, USBOR and EWEB will assess factors affecting methylation rates in their reservoirs by evaluating DEQ specified metrics. These metrics include (1) a reservoir specific- mercury translator, which relates water column total mercury to dissolved methylmercury, like the translator used in the TMDL model, (2) nutrient status, (3) dissolved oxygen profile, (4) water level fluctuations and (5) area of reservoir-adjacent wetlands affected by water level fluctuations. This assessment step will establish baseline conditions for use in adaptive management and inform evaluations of site-specific approaches to reduce methylmercury production. The DMAs/responsible persons will also identify specific measurable objectives with milestones and associated implementation timeline for implementing best management practices that address methylation rates in their reservoirs.

A TMDL implementation plan must be submitted to DEQ within 18 months of TMDL issuance. The plan will describe the timeline for completing the assessment of factors affecting methylation rates, evaluation of site-specific best management practices for reducing methylation, and implementing best management practices to address methylation rates in their reservoirs. See example best management practices in Table 13-19 for practices that could be included in implementation plans. The plan will also include a rationale for identifying specific measurable objectives and any additional DMA/responsible persons determined metrics used for tracking measurable objectives. Development of implementation plan elements for the Cottage Grove Reservoir must be coordinated with EPA’s Black Butte Mine Superfund Remedial Investigation and Feasibility Study.

The USACE, PGE, USBOR and EWEB will also take part in the Willamette Basin five year review. For more information about five year reviews, see Section 13.4.

U.S. Army Corps of Engineers Stream flow in the Willamette Basin is highly modified by dam and reservoir operations. The U.S. Congress passed 15 flood control acts between 1938 and 1974 that affect the Willamette Basin and are implemented by USACE. The 13 USACE dams comprise 91 percent of the total dam storage capacity in the basin. These dams provide flood control, navigation, hydroelectric power, and water in summer for irrigation, recreation, and downstream water quality. Dam operations have dramatically changed the natural flow patterns of the Willamette River by reducing peak flows in winter and artificially augmenting summer low flows.

Oregon Department of Environmental Quality 116 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

U.S. Bureau of Reclamation USBOR operates Scoggins Dam, which impounds Scoggins Creek forming Hagg Lake in the Tualatin sub-basin. Hagg Lake comprises approximately 2 percent of the total dam storage capacity in the Willamette Basin.

Eugene Water and Electric Board EWEB is Oregon’s largest customer-owned public utility providing electricity and water to Eugene and portions of East Springfield and the McKenzie River Valley. EWEB owns and operates Carmen Diversion, Smith River, Trail Bridge Reservoir, Leaburg Dam and Walterville Canal in the McKenzie sub-basin. The four dams comprise approximately 0.6 percent of the total dam storage capacity in the Willamette Basin. Water for the Walterville canal is diverted from the mainstem McKenzie River for the purposes of hydroelectric generation - there is no reservoir storage.

The Leaburg-Walterville Hydroelectric Project is comprised of two run-of-the-river dams on the McKenzie River. The Leaburg Dam impounds and diverts the McKenzie River through the Leaburg Canal to the Leaburg power plant. Flow from the Leaburg power plant returns to the McKenzie River. The impoundment forms the Leaburg Reservoir.

Portland General Electric PGE provides electricity to Portland, Salem and the surrounding areas. PGE owns and operates Timothy Lake, North Fork Dam, River Mill Dam, Faraday Forebay, Faraday Diversion Dam, Frog Lake and Harriet Lake. These seven dams comprise approximately 5 percent of the total dam storage capacity in the Willamette Basin.

13.3.2 Management strategies for point sources As required in OAR 340-042-0040(4)(l)(E), the following section describes management strategies for point sources. As noted in this TMDL, point source wasteload allocations are applied as percent reductions aggregated across two sectors – permitted wastewater discharges and permitted stormwater discharges. Wasteload allocations are assigned to the permitted source sectors, not to specific dischargers. DEQ determined that the most effective way to optimize mercury reductions is to apply mercury and erosion minimization and control measures that are appropriate for each sector, facility, land use, or activity. Reasonable assurance that point source wasteload allocations will be met is addressed through the issuance or revision of National Pollutant Discharge Elimination System permits.

13.3.2.1 NPDES Wastewater Permits As described in Section 10, the wastewater sector wasteload allocation is a 10 percent reduction from estimated existing mercury loads discharged under all wastewater permits. Permit categories under the aggregate 10 percent reduction wasteload allocation include: major and minor domestic sewage treatment plant permits; major and minor industrial wastewater permits; and wastewater discharges covered under non-stormwater general permits.

Oregon Department of Environmental Quality 117 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Domestic Sewage Treatment Plant Wastewater Permits Major sewage treatment plant facilities Major sewage treatment plant facilities are facilities with discharges greater than 1 million gallons per day, populations greater than 10,000 or with pretreatment programs classified as “major” and are listed in Table 9-3 in Section 9 on source assessment. For these major STP facilities, consideration of permit renewal will include enforceable conditions for monitoring and reporting of total mercury and development and implementation of mercury minimization programs, in accordance with the most recent version of DEQ’s Internal Management Directive on Implementation of Methylmercury Criterion in NPDES Permits, 2013. Required elements include: • Identification of potential sources of mercury in discharge; • Implementation and tracking of source reduction activities; • Monitoring to document effectiveness; and • Reporting. As part of the Accountability Framework described in Section 14.1, reporting from major STPs will be tracked and evaluated for progress toward the 10 percent overall wastewater sector reduction of approximately 0.44 g/day or 0.16 kg/yr. Minor sewage treatment plant facilities Within the Willamette Basin, estimated total discharge flows from all minor STPs are less than 10 percent of the total discharge flows from all major STPs. In the TMDL Technical Support Document (TetraTech, 2019a), the total mercury load from all minor STPs was estimated at 0.095 kilograms/year, or essentially 0 percent of the total mercury load in the basin. DEQ determined that the potential mercury load from minor STP discharges is a very minor contribution to the estimated 0.8 percent of total mercury load from all STPs within the basin. Therefore, no additional controls or monitoring will be required from minor STPs toward achieving the 10 percent overall wastewater sector reduction of 0.44 g/day or 0.16 kg/yr. As minors qualify to become majors, permit requirements will reflect those described above for major STPs.

Industrial and General Wastewater Permits As described in the TMDL Technical Support Document (TetraTech, 2019a), the following NPDES permitted industrial activity categories have the potential to include mercury in their process operations: • timber products; • paper products; • chemical products; • glass/clay/cement/concrete/gypsum products; • primary metal industries; • fabricated metal products; • electronics and instruments. Permits for facilities that do not include process wastewater discharges of any of the categories of activities in the list above will not include requirements specific to achieving a portion of the aggregated sector-specific 10 percent reduction wasteload allocation.

Oregon Department of Environmental Quality 118 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Major and minor industrial DEQ evaluated whether the existing eight major (as determined using EPA Industrial Classification worksheet) and 57 minor industrial wastewater permits in the Willamette Basin discharge process wastewater from any of the above categories of activities. DEQ determined that there are seven major industrial discharges with active permits, and 15 minor facilities that fall into SIC code categories with activities that have the potential to increase mercury in process wastewater discharge (see Table 9-4). DEQ will confirm these determinations during renewal of each permit. For confirmed facilities, DEQ will evaluate existing data to determine the significance of mercury loads in discharges. DEQ will also consider the potential for measurable reductions toward the 10 percent sector aggregate wasteload allocation in making a determination as to whether development and implementation of a mercury minimization plan is warranted for the facility. Depending on mercury and flow data availability and quality, permits will include, either:

1. If sufficient mercury and flow data exists, enforceable conditions for monitoring and reporting of influent and effluent total mercury and, if determined to be warranted, development and implementation of a mercury minimization plan, in accordance with the most recent version of DEQ’s Internal Management Directive on Implementation of Methylmercury Criterion in NPDES Permits, 2013. Required elements include:

• Identification of potential sources of mercury in discharge; • Implementation and tracking of source reduction activities; • Monitoring to document effectiveness; and • Reporting.

2. If there is insufficient mercury and flow data, enforceable conditions on influent and effluent monitoring and reporting of total mercury and discharge flows. After two years of data collection, effluent mercury and total suspended solids concentrations and discharge flows will be evaluated to determine estimated mercury load discharged, to determine whether development and implementation of a mercury minimization plan is warranted for the facility. Mercury influent data will also be evaluated in comparison to effluent to inform decisions regarding the need for mercury minimization plans.

As part of the Accountability Framework described in Section 14.1, reporting from these industrial facilities will be tracked and evaluated for progress toward the 10 percent overall wastewater sector reduction of approximately 0.44 g/day or 0.16 kg/yr. General wastewater With the exception of the 700PM general permit for suction dredge mining, DEQ determined that all categories of the 158 entities currently issued general wastewater permits (36 cooling water, 24 filter backwash, 4 fish hatcheries, 4 boiler blowdown, 9 petroleum hydrocarbon cleanup, 21 wash water, 60 pesticide application) have little to no potential for mercury to be increased in permitted discharges. In addition, flow volumes are insignificant as contributors to the estimated 0.3 percent total load of mercury from industrial discharges into Willamette basin streams. Therefore, no permit requirements are necessary specific to achieving a portion of the aggregated sector-specific 10 percent reduction wasteload allocation.

Oregon Department of Environmental Quality 119 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Discharge flows from suction dredges permitted under the 700PM generally are also insignificant. However, as noted in Section 9.4.1, when operated in areas of historical mercury contamination, studies in Oregon, California, Wisconsin and Florida have shown that significant levels of mercury can be disturbed, mobilized and methylated by suction dredging. The high potential for high concentrations of mercury to be released and converted in this specific subbasin constitutes a significant mercury load. Therefore, upon renewal of the 700PM permit, DEQ will prohibit dredging locations in streams that flow from the former Bohemia Mining District and are tributary to the Dorena Reservoir (including Row River, Brice Creek, Sharps Creek and Champion Creek).

Reductions from ceasing these discharges are expected to contribute to the 10 percent overall wastewater wasteload allocation of approximately 0.44 g/day or 0.16 kg/yr, but will be locally effective in Dorena Reservoir and its tributaries. This small portion of the wasteload allocation will be evaluated as part of the Monitoring Framework, being developed by DEQ and EPA.

13.3.2.2 NPDES Permitted Stormwater The permitted stormwater sector wasteload allocation is a 75 percent reduction from estimated existing mercury loads discharged under all stormwater permits. As noted in the TMDL and TMDL Technical Support Document (TetraTech, 2019a), atmospheric deposition is the major source of mercury. Once mercury is deposited on the landscape it can be eroded and/or transported in stormwater to rivers, streams and other waterbodies.

Permittees will be responsible for applying controls to prevent or reduce mercury discharges from within their jurisdictions in light of these mixed sources and delivery mechanisms of mercury. Controls cannot accurately distinguish or specifically target sources, thus DEQ acknowledges that some portion of background sources will be captured by permittee implemented controls and that some portion of sources will remain uncontrolled. The goal is to show achievement of measureable objectives within each jurisdiction toward a 75 percent reduction as an overall sector. Permit categories under the 75 percent reduction wasteload allocation include: MS4 Phase I; MS4 Phase II; 1200-A; 1200-Z; and1200-C/CN/CA.

Municipal Separate Stormwater Sewer System As noted in Section 9.4.2, coverage is required for 47 entities under Phase I and Phase II MS4 permits within the Willamette Basin, as listed in Table 9-5.

MS4 Phase I Upon permit renewal, each MS4 Phase I permit will include the following requirements: • Develop and submit a mercury minimization assessment with the second annual report of the renewed permit term, that documents the current actions, such as best management practices implemented, that reduce the amount of solids discharged into the permitted MS4 system (similar to the actions currently required in Schedule A of the permits). • Continued implementation of the best management practices and other actions described in the mercury minimization assessment that are effective for mercury reduction, along with documentation of implementation in each subsequent annual report . • An analysis of the effectiveness of the best management practices and any other actions taken and qualitative pollutant load reductions achieved in the fourth annual report. Due

Oregon Department of Environmental Quality 120 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

to data limitations, mercury benchmarks are not applicable in the first permit cycle after the TMDL is finalized. • Collection of paired total mercury and total suspended solids samples. • Submittal of paired mercury and total suspended solids monitoring data in the appropriate DEQ data submission template, and pollutant load reduction evaluations and wasteload allocation attainment analyses with mercury specific information.

MS4 Phase II DEQ’s MS4 Phase II general permit became effective in March 2019. The permit includes requirements for controlling erosion and other pollutants associated with solids entrained in stormwater. Therefore, the jurisdictions covered under the Phase II general permit will not be required to implement any additional control measures toward achieving the 75 percent reduction sector wasteload allocation during the permit term.

For Phase II jurisdictions covered under an individual permit, upon renewal each permit will include, at minimum, the conditions in the MS4 Phase II general permit effective at the time regarding construction and post-construction requirements or requirement to develop, submit and implement a mercury minimization plan with the goal of demonstrating achievement of objectives toward attaining the 75 percent overall sector reduction. The plan must include: • A description of both structural and non-structural control measures the permittee intends to implement; • An evaluation of current structural and non-structural control measures and their relative effectiveness; • A control measure effectiveness evaluation strategy to inform implementation of future control measures.

As part of the Accountability Framework described in Section 14.1, reporting from these MS4 Phase I and II jurisdictions will be tracked and evaluated for progress toward the 75 percent overall stormwater sector reduction of approximately 8.48 g/day or 3.10 kg/yr. DEQ will use information from the first permit cycles following issuance of the TMDL to determine future permit requirements, including consideration of paired TSS and mercury samples from MS4 Phase II permittees, that may be needed to adaptively manage mercury reduction achievement.

Stormwater General Permits (1200-A, 1200-Z and 1200-C/CN/CA) Most of the general stormwater permitted sites are located within MS4-permitted and non- permitted urban areas. In the Willamette Basin, these include approximately: 109 registrants under the 1200-A for non-metallic mining and asphalt and concrete plants; 629 registrants under the 1200-Z for industrial facilities; and approximately 1,000 short term registrants under the 1200-C/CN/CA for stormwater control during construction activities.

As noted in the TMDL and Technical Support Document (TetraTech, 2019a), mercury loads from general stormwater permits (1200-A, 1200-Z, and 1200-C/CN/CA) were implicit in the modeled MS4-permitted mercury load estimates. There are several existing requirements and planned revisions for these permits that DEQ expects will result in reduction of mercury loads contributing toward the achievement of the overall stormwater sector wasteload allocation of 75 percent reduction.

Oregon Department of Environmental Quality 121 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

The NPDES 1200-Z Industrial General Stormwater Permit was re-issued in 2017 and updated in 2018. The 1200-Z permit includes a reduced benchmark for total suspended solids for discharges into the geographic regions of the Portland Harbor (approximately the lowest 10 miles of the Willamette River) and the Columbia Slough. These are the most densely industrialized areas of the Willamette Basin and, according to the TMDL modeling, represent key areas for mercury load reductions from stormwater (TetraTech, 2019a). The total suspended solids benchmark for discharges to these areas was set at 30 mg/L, reduced from 100 mg/L for discharges into Portland Harbor and from 50 mg/L in the Columbia Slough. In part, the reduced benchmark targets reduction of toxic substances (including mercury) that are associated with solids in stormwater and wastewater discharges. Upon renewal, it is expected that the 1200-A permit will also include the 30 mg/L total suspended solids benchmark in these two key geographic areas. Implementation of the lowered total suspended solids benchmark in these permits, as well as prohibitions on turbid discharge in the widespread, but temporary 1200-C/CN/CA permits, is anticipated to enhance reduction of mercury loads toward achievement of the overall stormwater sector wasteload allocation of 75 percent reduction. As a result, mercury reductions achieved through current and future general stormwater permit requirements for permitted activities conducted within the MS4-permitted jurisdictions will contribute to the aggregate stormwater sector reductions needed to achieve the wasteload allocation. 13.3.3 Other DEQ Mercury Reduction Programs 13.3.3.1 Regulatory Programs Air Emissions Mercury Reductions DEQ achieves mercury reductions from air emissions through implementation of federal Title V permits, state Air Contaminant Discharge permits and the newly adopted state Cleaner Air Oregon program. On August 13, 2019, the state of Oregon joined 20 other states in a lawsuit against EPA’s decision to ease restrictions on coal-fired power plants. Given that emissions from coal-fired power plants across the country could contribute to mercury emissions nationwide, DEQ will continue to support regulatory efforts at the national level to reduce mercury emissions. Environmental Cleanup Program DEQ requires responsible parties to remediate contaminated land, groundwater and stream sediment as authorized by OAR 340-122-0070. DEQ Cleanup Program activities related to mercury are focused on abandoned mines in the state and responding with EPA to mercury spills. The Black Butte Mine site, which is a significant source of mercury to the Cottage Grove Reservoir, is an EPA Superfund site where cleanup actions were implemented in 2018 to address this source. State Legislation on Mercury in Products With regard to preventing mercury pollution, the Oregon Legislature adopted several bans, restrictions or management requirements for mercury in products since the 1990s. Those products include:

• Lighting fixtures • Novelty items • Thermostats, and • Vehicle switches

Oregon Department of Environmental Quality 122 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

In addition, the 2007 Legislature required dental offices to install dental amalgam separators and related maintenance best management practices to ensure mercury-containing amalgam waste does not end up in wastewater systems.

The state legislature passed Oregon’s Environmental Protection Act in 2019, which ensures that the federal environmental standards of the Clean Air and Clean Water Acts that were in place and effective as of January 19, 2017, will remain in effect and be enforceable under state law even if the federal government rolls back regulations.

13.3.3.2 Voluntary programs Household and small business mercury waste collection activities DEQ’s Solid and Hazardous Waste programs have initiated and implemented multiple specialized collection and exchange projects for mercury-containing products, including collecting mercury wastes at numerous one-day household hazardous waste collection events throughout Oregon. For more information about household hazardous waste events visit DEQ’s website: https://www.oregon.gov/deq/Hazards-and-Cleanup/hw/Pages/hhw.aspx. • Thermometers – A thermometer exchange program was initiated to reduce the amount of mercury in homes and ensure proper disposal of mercury thermometers. DEQ provides free digital thermometers at collection events to citizens turning in a mercury containing thermometer. DEQ also supplies local governments with free digital thermometers to encourage them to implement their own exchange programs. Currently, DEQ averages approximately one digital thermometer exchange for every 50 participants. • Thermostats – The Thermostat Recycling Incentive project was initiated by DEQ, Portland General Electric, the Thermostat Recycling Corporation and the Product Stewardship Institute to encourage recycling of mercury containing thermostats. Between 2006 and 2007, contactors participating in the program received $4 rebate coupons for each mercury-containing thermostat they returned to a participating wholesaler for recycling. The coupons could be used toward the purchase of mercury-free Energy Star ® qualified thermostats. From 2010 to 2013, DEQ covered the $25 registration cost for contractors and local governments to receive a Thermostat Recycling Corporation collection bin. • Dairy Manometers – DEQ worked with dairy and agricultural organizations in 2005 and 2006 to replace mercury manometers (pressure-measuring devices) used in dairy farm milking operations with mercury-free digital vacuum gauges. The mercury-containing manometers were managed and disposed of properly by DEQ’s hazardous waste contractor. An EPA grant provided $300 to each participant to cover most of the costs associated with supplying and installing the mercury-free replacement pressure- measuring device. • Dental Mercury Wastes – DEQ has been working with the Oregon Dental Association and the Oregon Association of Clean Water Agencies since 2003 to improve the management of mercury-containing wastes, such as dental amalgam. The partners sponsor an annual mercury waste collection event held in conjunction with the annual dental association conference. DEQ’s Solid Waste program funded the collection and disposal of the waste in collaboration with local household hazardous waste programs. • Mercury Auto Switches – The Northwest Auto Trades Association, the Oregon Environmental Council, local governments, and DEQ began a program in 2001 to replace mercury-containing automotive light switches in consumer automobiles with

Oregon Department of Environmental Quality 123 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

mercury-free ball-bearing switches free of charge. Eligible cars were 2002 and older. DEQ’s Hazardous Waste program also developed and distributed a fact sheet on mercury switch removal for automobile dismantlers in Oregon. • Suction Dredge Mining Waste Mercury – DEQ worked with a hobby mining association in 2002 and 2003 on various activities including sponsoring two mercury waste collection events in Myrtle Creek. • Fluorescent Lamps – Fluorescent light tubes and compact fluorescent bulbs can be taken to a household hazardous waste collection event or facility. For more information about collection events visit DEQ’s website: https://www.oregon.gov/deq/Hazards-and- Cleanup/hw/Pages/Mercury-Disposal.aspx

Household and small business mercury education and reporting activities DEQ’s Solid and Hazardous Waste programs continue to partner with various organizations, local governments and non-profits to educate households and businesses about proper management of mercury-containing products and alternatives. DEQ also initiated an effort to collect better data on mercury waste generated by businesses. Specific activities implemented between 2002 and 2006 include the following: • Educational materials – DEQ developed educational fact sheets on the proper management of mercury-containing products and wastes, including cleaning up mercury spills. • Dental offices – At the Oregon Dental Association’s annual conference DEQ staff assist with educational outreach to participating dentists. In addition, DEQ developed a simplified tax credit application and fact sheet for dentists installing amalgam separators. • Fluorescent lamps – The Hazardous Waste program participated in several lighting fairs sponsored by electric utilities to provide educational information on proper disposal of mercury-containing fluorescent lamps. In addition, DEQ worked with the Oregon Environmental Council to develop a lamp fact sheet for property management companies. • Suction dredge miners – DEQ developed printed educational information for miners on proper mercury management • Reporting on mercury containing hazardous waste –DEQ’s hazardous waste generation annual reporting form was modified to request specific information on the generation and management of mercury containing wastes from businesses and other entities required to submit these reporting forms.

13.4 Timeline for implementing management strategies The purpose of this element of the WQMP, required by OAR 340-042-0040(4)(l)(D), is to demonstrate a strategy for implementing and maintaining the implementation plan, and to evaluate water quality improvements over time. Included in this section are timelines for TMDL implementation activities for nonpoint sources and point sources.

Oregon Department of Environmental Quality 124 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Table 13-24. The timeline for activities related to this WQMP and associated DMA and responsible person Implementation plans, and NPDES permits. Activity Year of Activity 2019 2020 2021 2022 2023 DEQ modification of affected NPDES Wastewater and Upon permit renewal Stormwater Permits Ongoing implementation of DEQ- approved plans that X X DMAs and responsible persons already have in place Designated Management Agencies and responsible persons (see Appendix E of WQMP) develop and/ or X X update, and submit implementation plans within 18 months of TMDL issuance Implementation of new, updated or revised DMA and X X X responsible person implementation plans X DMA and responsible person submittal of annual reports X X X X

DEQ, DMA and responsible person five year review of X

implementation

13.4.1 Nonpoint Source DMAs and responsible persons Each nonpoint source DMA and responsible person will submit a new or revised TMDL implementation plan that includes timelines for implementation of the measurable objectives and milestones described in Section 13.3.1. Timelines will be specific wherever possible and will include a schedule for implementation and evaluation of strategies, and reporting dates and milestones for evaluating progress. TMDL implementation plans must be submitted to DEQ for approval within 18 months of the issuance of the TMDL, or earlier if desired (for example, DMA’s may wish to have their plans coincide with already established deadlines for annual reports). DMAs should work with DEQ basin coordinators on specific submission requirements.

Adaptive management is a central element of individual implementation plans, this WQMP, and the TMDL. As part of adaptive management, DEQ intends to regularly review the progress of implementation plans. Through ongoing monitoring and evaluation, DEQ, DMAs and responsible persons can learn from experience and modify policy and implementation approaches in order to achieve better environmental outcomes.

Annual reports Cities, counties and some special districts that have been named DMAs in this WQMP will have annual reporting requirements. DMAs will report on progress in implementing nonpoint source strategies identified in the TMDL implementation plans, including any delays or challenges DMAs had in implementing strategies. DMAs may combine reporting for mercury along with other Willamette Basin TMDL pollutants. Annual reports must be posted on a publicly accessible website unless a DMA does not have a website.

Responsible persons and DMAs (which include some special districts, and local, state and federal agencies) will report on implementation progress of nonpoint source strategies, which may include annual reports. Implementation strategies will be identified in TMDL implementation plans, as described in an existing Memorandum of Understanding or Memorandum of Agreement, or as directed by DEQ.

Oregon Department of Environmental Quality 125 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Willamette Basin TMDL Five Year Review The 2006 Willamette Basin TMDL required the development and submission of TMDL implementation plans with annual reporting to DEQ. The 2006 TMDL also required DMAs and responsible persons to submit a report every five years to assess effectiveness of the management strategies identified in implementation plans and in place during the preceding four years. As part of the five year review, DEQ evaluates the number of implementation plans and annual reports submitted by DMAs and responsible persons, and the adequacy of the strategies contained in those plans to reduce pollutant inputs and restore water quality. These reviews have provided valuable feedback to the agency on successes and challenges DMAs experience in implementing their nonpoint source program. For this reason, DEQ will continue to require all nonpoint source DMAs and responsible persons, unless otherwise notified by DEQ, to include progress in implementation of mercury reduction strategies with their five year report as described in this WQMP.

Willamette Basin five year reviews occurred in 2013 and 2018, and the Molalla-Pudding five year review occurred in 2015. The next five-year reviews for the Molalla-Pudding and the Willamette Basin TMDLs are planned to occur in 2021 and 2023, respectively. DEQ expects that management strategies related to mercury will be included in the Willamette Basin 2023 five year review, even though four complete years of mercury implementation based on the updated WQMP will not have occurred by then. The objective of this timeline is to retain a consistent five-year reporting cycle for current and future Willamette Basin TMDLs. DEQ will post five year review reports to its website.

In the five year reviews, DMAs and responsible persons must address progress in implementing mercury reduction strategies, in addition to other nonpoint source pollutants established under previous Willamette TMDLs for which they were named as DMAs or responsible persons. Details of this submittal will be provided by DEQ to DMAs and responsible persons in advance of the deadline for these reports. Entities such as state and federal agencies with a Memorandum of Understanding or Memorandum of Agreement with DEQ may have different or additional reporting requirements.

During the five year review, DMAs must review their implementation plans in collaboration with DEQ staff to evaluate whether strategies, timelines, milestones, or other components of the plan should be updated for the next five years. DMAs and responsible persons may also update implementation plans more often than every five years due to significant changes in TMDL pollutant reduction strategies or program priorities. 13.4.2 Point sources Provisions to address the appropriate point source wasteload allocations will be incorporated into National Pollutant Discharge Elimination System permits when permits are renewed by DEQ. A schedule for meeting the requirements associated with this TMDL will be incorporated into the permit. Like other permit conditions, compliance with the terms and conditions of the permit is required by state and federal law. NPDES permittees will implement the permit renewal requirements described in Section 13.3.2.

Oregon Department of Environmental Quality 126 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

13.5 Timeline for attainment of water quality standards This WQMP component is required by OAR 340-042-0040(4)(l)(F). The timeline for attainment of water quality standards for this TMDL is expected to take multiple decades. The primary source of mercury in the basin is air deposition, and while efforts to reduce emissions in North America are ongoing, continued air emissions from global sources may offset these efforts. Other sources of mercury are varied and include buffering and re-release of mercury from the ocean, re-suspension of sediment-bound mercury in waterbodies, and changes in total mercury in groundwater. These legacy mercury deposits will take years to diminish.

Nonpoint sources of mercury contribute more mercury to the basin relative to point sources. Therefore, it is especially important for this TMDL for nonpoint sources to make timely progress toward meeting the TMDL load allocations. DEQ expects nonpoint source DMAs and responsible persons to meet the interim milestones for percent reductions (Table 13-25). If interim milestones are not met, DEQ may require DMAs and responsible persons to revise their implementation plans and implementation timelines accordingly (OAR 340-042-0080(4)(b)).

If DEQ determines that non-federal forest operations regulated under the Forest Practices Act are not making satisfactory progress toward meeting milestones or achieving load allocations, or if DEQ determines that the general Forest Practices Act rules are not sufficient for meeting allocations, site specific rules under the Forest Practices Act rules will need to be created or revised. If the site specific rules are not implemented, DEQ will request the Environmental Quality Commission to petition the Board of Forestry to make necessary changes (OAR 340- 042-0080(2)).

If DEQ determines that agricultural practices subject to the Agricultural Water Quality Management Act are not making satisfactory progress toward meeting milestones or achieving load allocations, or if the area plan and rules are not adequate to ensure implementation of the load allocation, the department will provide Oregon Department of Agriculture with comments on what would be sufficient to meet TMDL load allocations during each biennial review process. Should that effort not be sufficient DEQ will request the Environmental Quality Commission to petition ODA to make the necessary changes (OAR 340-042-0080(3)).

Table 13-25. Timeline for reaching interim milestones for the general nonpoint source 88 percent reduction in instream mercury levels. Assessment of progress will be supported by water quality monitoring conducted by DEQ and watershed partners. Assessment Year Cumulative Percent Reduction Milestones for Instream Mercury 2028 30 2038 60 2048 88 13.6 Monitoring and evaluation As required in OAR 340-042-0040(4)(l)(K), this section describes DEQ’s plan to monitor and evaluate progress toward achieving TMDL allocations and water quality standards.

Accountability and evaluation has two basic components: 1) tracking the implementation of DMA-specific water quality implementation plans identified in this document and 2) monitoring

Oregon Department of Environmental Quality 127 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

the physical, chemical and biological parameters for water quality. Monitoring will provide a check on the progress being made toward achieving the TMDL allocations and meeting water quality standards, and will be used as part of the Adaptive Management process (Figure 13-2) The estimated timeline for achieving water quality standards is multiple decades.

The objectives of this monitoring effort are to demonstrate long-term recovery, better understand natural variability, and track implementation of projects and best management practices, and track effectiveness of TMDL implementation. This monitoring and feedback mechanism is a major component of the “reasonable assurance of implementation” for the Willamette Basin WQMP.

DMA-specific implementation plans will be tracked by accounting for the numbers, types and locations of projects, best management practices, education activities, or other actions taken to improve or protect water quality. The mechanism for tracking DMA and responsible person implementation efforts will be annual reports to be submitted to DEQ.

The information generated by each of the agencies or entities gathering data in the Willamette Basin will be pooled and used to determine whether management actions are having the desired effects or if changes in management actions and/ or TMDLs are needed. This detailed evaluation will typically occur on a five year cycle. If progress is not occurring, then the appropriate DMA or responsible person will be contacted with a request for action.

DEQ and EPA are currently developing an Assessment and Monitoring Strategy to Support Implementation of Mercury Total Maximum Daily Loads for the Willamette Basin. This monitoring strategy will be used to evaluate effectiveness of DMA and responsible person implementation strategies at meeting allocations and may require certain DMAs to collect data. The monitoring strategy will also be used to determine progress in the Willamette River and its tributaries toward meeting the total mercury loading capacity of 0.14 ng/L, methylmercury fish tissue criterion of 0.040 mg/kg, and instream total suspended solid surrogate allocations. DEQ will involve stakeholders in the process of developing the monitoring strategy. DEQ will finalize this monitoring strategy after the issuance of the TMDL. Figure 13-6. Multnomah Channel, Lower Willamette. 13.7 Costs and funding This section provides a general discussion of costs and funding for implementing management strategies as required by Oregon Administrative Rule 340-042-0040(4)(l)(N). Please note that sector-specific or source-specific implementation plans may provide more detailed analyses of costs and funding for specific management strategies.

Designated Management Agencies will be expected to provide a fiscal analysis of the resources needed to develop, execute and maintain the programs described in their Implementation plans.

Oregon Department of Environmental Quality 128 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

The purpose of this element is to describe estimated costs and demonstrate there is sufficient funding available to begin implementation of the WQMP. Another purpose is to identify potential future funding sources for project implementation.

Generally DMAs should estimate funding needed over a five-year timeframe to coincide with implementation strategies identified in its TMDL implementation plan. Factors to consider include: . Staff salaries, supplies, volunteer coordination, regulatory fees . Installation, operation, and maintenance of management measures . Monitoring, data analysis and management . Education and outreach efforts . Ordinance development

Funding is essential to implementing projects associated with this WQMP. There are many sources of local, state, and federal funds. Table 13-26 provides a partial list of funding and assistance programs available in the Willamette Basin.

Table 13-26. Partial list of funding programs available in the Willamette Basin that may be used to support planning and implementation activities that benefit water quality Program General Description Contact Loan program for below-market rate loans for planning, Clean Water State design, and construction of various water pollution control DEQ Revolving Fund activities. Conservation Provides annual rent to landowners who enroll Reserve agricultural lands along streams. Also cost-shares NRCS, SWCDs, Enhancement conservation practices such as riparian tree planting, ODF Program (CREP) livestock watering facilities, and riparian fencing. Competitive CRP provides annual rent to landowners Conservation who enroll highly erodible lands. Continuous CRP Reserve Program provides annual rent to landowners who enroll agricultural NRCS, SWCDs (CRP) lands along seasonal or perennial streams. Also cost- shares conservation practices such as riparian plantings. Provides cost-share and incentive payments to Conservation landowners who have attained a certain level of Stewardship Program NRCS, SWCDs stewardship and are willing to implement additional (CSP) conservation practices. These funds allow states to provide loans for certain Drinking Water source water assessment implementation activities, Oregon Health Source Protection including source water protection land acquisition and Authority Fund other types of incentive-based source water quality protection measures. Available through the USDA-Natural Resources Conservation Service. Provides federal funds for Emergency emergency protection measures to safeguard lives and Watershed Protection NRCS, SWCDs property from floods and the products of erosion created Program (EWP) by natural disasters that cause a sudden impairment to a watershed. Available through the USDA-Natural Resources Emergency Forest Conservation Service. Helps owners of non-industrial Restoration Program USDA, ODF private forests restore forest hrealth damaged by natural (EFRP) disasters. Environmental Fund projects that improve watershed functions and DEQ, SWCDs, Protection Agency protect the quality of surface and groundwater, including Watershed Section 319 Grants restoration and education projects. Councils

Oregon Department of Environmental Quality 129 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Program General Description Contact Cost-shares water quality and wildlife habitat Environmental improvement activities, including conservation tillage, Quality Incentives NRCS, SWCDs nutrient and manure management, fish habitat Program (EQIP). improvements, and riparian plantings. Farm and Ranchland Cost-shares purchases of agricultural conservation NRCS, SWCDs, Protection Program easements to protect agricultural land from development. ODF (FRPP) Federal Reforestation Internal Revenue Provides federal tax credit as incentive to plant trees. Tax Credit Service NRCS, Farm Grassland Reserve Provides incentives to landowners to protect and restore Service Agency, Program (GRP) pastureland, rangeland, and certain other grasslands. SWCDs U.S. Fish and Landowner Incentive Provides funds to enhance existing incentive programs Wildlife Service, Program (LIP) for fish and wildlife habitat improvements. ODFW Provides grants for a variety of restoration, assessment, Oregon Watershed SWCDs, monitoring, and education projects, as well as watershed Enhancement Board Watershed council staff support. 25 percent local match requirement (OWEB) Councils, OWEB on all grants. Oregon Watershed SWCDs, Provides grants up to $10,000 for priority watershed Enhancement Board Watershed enhancement projects identified by local focus group. Small Grant Program Councils, OWEB Provides financial and technical assistance to private and non-federal landowners to restore and improve wetlands, U.S. Fish and Partners for Wildlife riparian areas, and upland habitats in partnership with the Wildlife Service, Program U.S. Fish and Wildlife Service and other cooperating NRCS, SWCDs groups. Program available to state agencies and other eligible organizations for planning and implementing watershed Public Law 566 improvement and management projects. Projects should NRCS, SWCDs Watershed Program reduce erosion, siltation, and flooding; provide for agricultural water management; or improve fish and wildlife resources. Resource Resource Conservation & Provides assistance to organizations within RC & D areas Conservation and Development (RC & in accessing and managing grants. Development D) Grants Provides for reforestation of under-productive forestland State Forestation Tax not covered under the Oregon Forest Practices Act. ODF Credit Situations include brush and pasture conversions, fire damage areas, and insect and disease areas. Provides cost share dollars through USFS funds to family Stewardship Program ODF forest landowners to have management plans developed. State Tax Credit for Provides tax credit for part of the costs of voluntary fish Fish Habitat habitat improvements and required fish screening ODFW Improvements devices. Cost-sharing program for landowners to protect and Stewardship enhance forest resources. Eligible practices include tree NRCS, SWCDs, Incentive Program planting, site preparation, pre-commercial thinning, and ODF (SIP) wildlife habitat improvements. Wetlands Reserve Provides cost-sharing to landowners who restore NRCS, SWCDs Program (WRP) wetlands on agricultural lands. Wildlife Habitat Provides cost-share for wildlife habitat enhancement NRCS, SWCDs Incentives Program activities.

Oregon Department of Environmental Quality 130 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Program General Description Contact Maintains farm or forestry deferral for landowners who Wildlife Habitat Tax ODFW, SWCDs, develop a wildlife management plan with the approval of Deferral Program NRCS the Oregon Department of Fish and Wildlife. Funding Resources for Watershed EPA website containing numerous links to funding

Protection and sources Restoration 13.8 Citation legal authorities As required in Oregon Administrative Rule 340-042-0040(4)(l)(O), this section cites legal authorities relating to implementation of management strategies.

Clean Water Act, Section 303(d) The DEQ is the Oregon state agency responsible for implementing the Clean Water Act in Oregon. The EPA delegates many Clean Water Act authorities to the State of Oregon which is administered by the Oregon Environmental Quality Commission through Oregon Revised Statute. Section 303(d) of the 1972 Federal Clean Water Act as amended requires states to develop a list of rivers, streams and lakes that cannot meet water quality standards without application of additional pollution controls beyond the existing requirements on industrial sources and sewage treatment plants. These waters are referred to as “water quality limited.” Water quality limited waterbodies must be identified by the EPA or by a state agency which has been delegated this responsibility by EPA. In Oregon, the responsibility to delegate water quality limited waterbodies rests with DEQ and DEQ’s list of water quality limited waters is updated every two years. The list is referred to as the 303(d) list. Section 303 of the Clean Water Act further requires that TMDLs be developed for all waters on the 303(d) list. The Oregon Environmental Quality Commission granted the DEQ Director authority to develop TMDLs and issue them as orders (OAR 340-042-0060). DEQ was granted authority by the commission to implement TMDLs through OAR 340-042 with special provisions for agricultural lands and nonfederal forestland as governed by the Agriculture Water Quality Management Act and the Forest Practices Act, respectively. The EPA has the authority under the Clean Water Act to approve or disapprove TMDLs that states submit. When a TMDL is officially submitted by a state to EPA, EPA has 30 days to take action on the TMDL. In the case where EPA disapproves a TMDL, EPA must issue a TMDL within 30 days. A TMDL defines the amount of pollution that can be present in the waterbody without causing water quality standards to be violated. A WQMP is developed to describe a strategy for reducing water pollution to the level of the load allocations and waste load allocations prescribed in the TMDL, which is designed to restore the water quality and result in compliance with the water quality standards. In this way, the designated beneficial uses of the water will be protected for all citizens.

Endangered Species Act, Section 6 Section 6 of the 1973 federal Endangered Species Act, as amended, encourages states to develop and maintain conservation programs for federally listed threatened and endangered species. In addition, Section 4(d) of the ESA requires the National Marine Fisheries Service to list the activities that could result in a “take” of species they are charged with protecting. With regard to this TMDL, NMFS’ protected species are salmonid fish. NMFS also described certain precautions that, if followed, would preclude prosecution for take even if a listed species were

Oregon Department of Environmental Quality 131 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

harmed inadvertently. Such a provision is called a limit on the take prohibition. The intent is to provide local governments and other entities greater certainty regarding their liability for take.

NMFS published their rule in response to Section 4(d) in July of 2000 (see 65 FR 42421, July 10, 2000). The NMFS 4(d) rule lists 12 criteria that will be used to determine whether a local program incorporates sufficient precautionary measures to adequately conserve fish. The rule provides for local jurisdictions to submit development ordinances for review by NMFS under one, several or all of the criteria. The criteria for the Municipal, Residential, Commercial and Industrial Development and Redevelopment limit are listed below: 1. Avoid inappropriate areas such as unstable slopes, wetlands, and areas of high habitat value; 2. Prevent stormwater discharge impacts on water quality; 3. Protect riparian areas; 4. Avoid stream crossings – whether by roads, utilities, or other linear development; 5. Protect historic stream meander patterns; 6. Protect wetlands, wetland buffers, and wetland function; 7. Preserve the ability of permanent and intermittent streams to pass peak flows (hydrologic capacity); 8. Stress landscaping with native vegetation; 9. Prevent erosion and sediment run-off during and after construction; 10. Ensure water supply demand can be met without affecting salmon needs; 11. Provide mechanisms for monitoring, enforcing, funding and implementing; and 12. Comply with all other state and federal environmental laws and permits.

Oregon Revised Statute Chapter 468B DEQ is authorized by law to prevent and abate water pollution within the State of Oregon. Particularly relevant provisions of this chapter include:

ORS 468B.020 Prevention of pollution (A) Pollution of any of the waters of the state is declared to be not a reasonable or natural use of such waters and to be contrary to the public policy of the State or Oregon, as set forth in ORS 468B.015. (B) In order to carry out the public policy set forth in ORS 468B.015, the Department of Environmental Quality shall take such action as is necessary for the prevention of new pollution and the abatement of existing pollution by: a) Fostering and encouraging the cooperation of the people, industry, cities and counties, in order to prevent, control and reduce pollution of the waters of the state; and b) Requiring the use of all available and reasonable methods necessary to achieve the purposes of ORS 468B.015 and to conform to the standards of water quality and purity established under ORS 468B.048.

ORS 468B.110 provides DEQ and the EQC with authority to take actions necessary to achieve and maintain water quality standards, including issuing TMDLs and establishing wasteload allocations and load allocations.

Oregon Department of Environmental Quality 132 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

NPDES and WPCF Permits DEQ administers two different types of wastewater permits in implementing Oregon Revised Statute (ORS) 468B.050. These are: the NPDES permits for waste discharge into waters of the United States; and Water Pollution Control Facilities permits for waste disposal on land. The NPDES permit is also a federal permit and is required under the Clean Water Act. The WPCF permit is a state program.

401 Water Quality Certification Section 401 of the CWA requires that any applicant for a federal license or permit to conduct any activity that may result in a discharge to waters of the state must provide the licensing or permitting agency a certificate from DEQ that the activity complies with water quality requirements and standards. These include certifications for hydroelectric projects and for ‘dredge and fill’ projects. The legal citations are: 33 U.S.C. 1341; ORS 468B.035 – 468B.047; and OAR 340-048-0005 – 340-048-0040.

USACE Dam Operation and Management In association with other federal statues, including House Document No. 531 Volume V, the River and Harbor Act, the Flood Control Act, and the Water Resources Development Act, the USACE is charged with operating its projects in compliance with the federal Clean Water Act, and in accordance with all federal, State, interstate and local requirements, administrative authority, and process and sanctions respecting the control and abatement of water quality pollution as per Title 1 Section 313 (33 U.S.C. 1323).

Oregon Forest Practices Act The Oregon Department of Forestry is the designated management agency for regulating land management actions on non-federal forestry lands that impact water quality (ORS 527.610 to 527.992, and OAR 629 Divisions 600 through 665). The Board of Forestry has adopted water protection rules, including but not limited to OAR Chapter 629, Divisions 625, 630, and 635-660, which describe best management practices for forest operations. The Oregon Environmental Quality Commission, Board of Forestry, DEQ, and ODF have agreed that these pollution control measures will primarily be relied upon to result in achievement of state water quality standards. Statutes and rules also include provisions for adaptive management that provide for revisions to FPA practices where necessary to meet water quality standards. These provisions are described in ORS 527.710, ORS 527.765, OAR 629-035-0100, and OAR 340-042-0080.

Agricultural Water Quality Management Act The Oregon Department of Agriculture is responsible for the prevention and control of water pollution from agricultural activities as directed and authorized through the Agricultural Water Quality Management Act, adopted by the Oregon legislature in 1993 (ORS 568.900 to ORS 568.933). It is the lead state agency for regulating agriculture for water quality (ORS 561.191).

The Agricultural Water Quality Management Plan Act directs the ODA to work with local communities to develop water quality management plans for specific watersheds that have been identified as violating water quality standards and have agriculture water pollution contributions. The agriculture water quality management plans are expected to identify problems in the watershed that need to be addressed and outline ways to correct the problems. Water Quality

Oregon Department of Environmental Quality 133 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

area rules for areas within the Willamette Basin include OAR 603-095-2100 to 1160, OAR 603- 095-2300 to 2360, OAR 603-095-2600 to 2660, and OAR 603-095-3700 to 3760. Local Ordinances Local governments are expected to describe in their Implementation plans their specific legal authorities to carry out the management strategies chosen to meet the TMDL allocations. Legal authority to enforce the provisions of a city’s NPDES permit would be a specific example of legal authority to carry out management strategies.

Oregon Department of Environmental Quality 134 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 14. Reasonable Assurance

OAR 340-042-0030(9) defines Reasonable Assurance as “a demonstration that a TMDL will be implemented by federal, state or local governments or individuals through regulatory or voluntary actions including management strategies or other controls.” OAR 340-042- 0040(4)(l)(J) requires a description of reasonable assurance that management strategies and sector-specific or source-specific implementation plans will be carried out through regulatory or voluntary actions.

The Clean Water Act section 303(d) requires that a TMDL be “established at a level necessary to implement the applicable water quality standard.” Federal regulations define a TMDL as “the sum of the individual wasteload allocations for point sources and load allocations for nonpoint sources and natural background” [40 CFR 130.2(i)].

When a TMDL is developed for waters impaired by point sources only, the existence of the NPDES regulatory program and the issuance of NPDES permits provide the reasonable assurance that the wasteload allocations in the TMDL will be achieved. That is because federal regulations implementing the Clean Water Act require that water quality-based effluent limits in permits be consistent with “the assumptions and requirements of any available [wasteload allocation]” in an approved TMDL [40 CFR 122.44(d)(1)(vii)(B)].

Where a TMDL is developed for waters impaired by both point and nonpoint sources, it is the state’s and EPA’s best professional judgment as to reasonable assurance that the TMDL’s load allocations will be achieved. EPA past practice directs that these determinations include consideration of whether practices capable of reducing the specified pollutant load: (1) exist; (2) are technically feasible at a level required to meet allocations; and (3) have a high likelihood of implementation.

Where there is a demonstration that nonpoint source load reductions can and will be achieved; a determination that reasonable assurance exists and, on the basis of that reasonable assurance, allocation of greater loads to point sources is appropriate. Without a demonstration of reasonable assurance that relied-upon nonpoint source reductions will occur, reductions to point sources wasteload allocations are needed.

Because of the well-documented lag time for instream responses to effective mercury reductions from controls on point and nonpoint sources (United Nations Environment Programme, 2019), DEQ anticipates that attainment of instream target mercury concentrations and reduced fish tissue methylmercury concentrations will take decades.

The Willamette Basin Mercury TMDL was developed to address both point and nonpoint sources with load reduction allocations proportional to estimated source contributions and in consideration of opportunities for effective measures to reduce those contributions. There are several elements that combine to provide the reasonable assurance to meet federal and state requirements. Education, outreach, technical and financial assistance, permit administration, permit enforcement, DMA or responsible person’s implementation and DEQ enforcement of TMDL implementation plans will all be used to ensure that the goals of this TMDL are met. Details of these elements are provided in the WQMP (Section 13) and are summarized in the sections that follow.

Oregon Department of Environmental Quality 135 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

14.1 Accountability framework Reasonable assurance that needed load reductions will be achieved for nonpoint sources is based primarily on an accountability framework incorporated into the WQMP, together with the implementation plans of DMAs and responsible persons. This approach is similar to the accountability framework adopted by EPA for the Chesapeake Bay TMDL, which was adopted in 2010 and can be accessed from EPA’s website: https://www.epa.gov/chesapeake-bay- tmdl/chesapeake-bay-tmdl-document. The reasonable assurance and accountability framework for this draft TMDL include the following elements:

Pollutant reduction strategies DEQ tracks Identify water relevant quality Designated status and Management trends Agencies Reasonable Assurance DEQ action Develop when timelines, Designated targets, Management measurable Agencies fail objectives to implement DEQ evaluates implementa- tion plans and progress

Figure 14-1. A Representation of the Reasonable Assurance Accountability Framework Led by DEQ.

14.1.1 Pollutant reduction strategies Section 13.3 identifies management strategies and specific implementation actions needed to achieve the identified pollutant reductions. These strategies and actions are comprehensively implemented through a variety of regulatory and non-regulatory programs. Many of these are existing strategies and actions that are already being implemented within the basin or elsewhere in the state and demonstrate reduced mercury loading. These strategies are technically feasible at an appropriate scale in order to meet the load allocations that are proposed for DMAs and responsible persons. A high likelihood of implementation is demonstrated because DEQ reviews the individual implementation plans and proposed actions for adequacy, establishes a monitoring and reporting system to track implementation and is establishing surrogate outcome measures that also will be monitored. Where implementation is not occurring, or where surrogate measures are not being met, DEQ will take action to require DMAs and responsible persons to take corrective action. Key reduction strategies include: control of all air emissions

Oregon Department of Environmental Quality 136 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

sources greater than 1 kg/year of mercury within Oregon; implementation of Oregon-wide dental amalgam treatment since 2007; Oregon bans on products containing mercury; remediation of legacy mining mercury sources; point source permit requirements; and nonpoint source implementation plans from 12 state and federal agencies and dozens of local governments, special districts and other responsible parties. 14.1.2 Identify relevant DMAs Section 13.2.3 and Appendix E: List of designated management agencies and responsible persons identify approximately 171 DMAs and responsible persons that will implement the WQMP management strategies and develop or revise their own implementation plan. This category captures additional entities identified since the 2006 TMDL was issued. In this 2019 revision, DEQ is including explicit allocations and requirements for control of mercury in stormwater and direct discharges from urban areas that do not yet meet the population thresholds requiring municipal stormwater permits and also for water conveyance maintenance practices. This significantly expands the numbers of DMAs and responsible persons actively applying mercury controls in the Willamette Basin. DEQ Willamette Basin coordinators work individually with these DMAs and responsible persons on developing and implementing the required management strategies to reduce mercury. All of these factors increases robustness of TMDL implementation throughout the basin. 14.1.3 Develop timeline, targets, measurable objectives Section 13.4 provides comprehensive timelines for implementing management strategies. This includes schedules for revising permits, submittal of reports, achieving appropriate incremental and measurable water quality targets, and completion of other measurable milestones. These timelines support the accountability framework by requiring timely action by both DEQ and DMAs and responsible persons so that enforcement and adaptive management actions can be triggered and evaluation of attainment of TMDL goals occurs. 14.1.4 Evaluate implementation plans and progress As provided in Section 13.4, DEQ will evaluate new or revised implementation plans from DMAs and responsible persons. This will ensure that the actions and measures included in the plans are feasible and have a high likelihood of being implemented and achieving load allocations. In addition, DEQ is proposing TSS as a surrogate measure for evaluating implementation of the allocations for the mainstem Willamette River and its tributaries. TSS will be used for evaluating the effectiveness of implementation plans. Monitoring locations will be described in the Assessment and Monitoring Strategy to Support Implementation of Mercury Total Maximum Daily Loads for the Willamette Basin. DEQ will use the monitoring data to determine trends in both the TSS surrogate measure and available data for mercury in the water column and in fish tissue.

As noted in Sections 13.5 and 13.6, DEQ will track the management strategies being implemented and evaluate achievements against established timelines and milestones. At a minimum, this will occur in the Willamette Basin through DEQ’s Five Year Reviews.

In making determinations about the effectiveness and implementability of mercury reduction measures, DEQ relies heavily on DMA and responsible person experience with measures specific to reducing erosion and runoff from their specific activities. The wide variety of potential actions that will be applied by 171 DMAs and responsible persons across dozens of point

Oregon Department of Environmental Quality 137 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 source sectors and land use activities prevent unilateral mandating of preferred practices and conclusions about their specific success. However, examples of where proven techniques are applied to reduce mercury give confidence that DEQ’s approach is reasonable and will effective for reducing mercury. Some examples of effective controls since implementation of the 2006 Willamette Basin Mercury TMDL began include: • Oregon’s two most significant air emissions mercury sources in 2006 were a coal-fired electric generation plant in Boardman and a cement plant in Durkee. In 2007, DEQ put strict control requirements in place on these facilities. Reductions in mercury emissions of 94 and 97 percent, respectively, have since been achieved and the coal plant is closing in 2020. • The 2019 Cleaner Air Oregon regulations will address the largest air source of mercury in the Willamette Basin. This source currently comprises 70 percent of the total mercury air emissions within the Willamette Basin. Controls under this program are expected to achieve significant reductions. • Clean Water Services operates four municipal sewage treatment plants serving more than a half a million residents in Washington County, Oregon. Advanced treatment technologies are employed at its facilities and mercury minimization measures have been implemented since at least 2004. While the systems are not designed specifically to address mercury, the facilities consistently achieve 97 to 99 percent mercury removal efficiencies. Effectiveness of mercury minimization measures, particularly reduced dental amalgam contributions, is also demonstrated by declining levels of mercury in biosolids between 2006 to 2018. • In the Lower Willamette subbasin, about a third of the City of Portland is served by a combined sewer system that carries sewage and some stormwater to the Columbia Boulevard Wastewater Treatment Plant for treatment before discharge to the Columbia River. The City of Portland completed controls on this system in phases between 2000, 2006 and 2011 to reduce combined sewage overflows to the Columbia Slough by 99% and to the Willamette River by 94%. Loads of sediment and runoff that potentially contain mercury have also been reduced. • Also in the Lower Willamette subbasin, DEQ completed four in-water sediment and bank remediation projects between 2011 and 2016. Removal, capping and bank stabilization at the following locations reduced mercury, along with other contaminants: Zidell at River Mile 13.5 to 14 in 2011 and 2012; Central District Greenway at River Mile 14.1 in 2013; and PGE properties at River Miles 13.1 and 13.5 in 2015 and 2016. In addition, other in- water remedial actions are planned and bank stabilization and significant stormwater control projects have been completed throughout the Portland Harbor reach (approximately River Mile 2-11). • ODA and DEQ have worked together to complete biennial reviews of Agricultural Water Quality Management Plans in the Upper, Middle and Lower Willamette Basin areas. These reviews report on water quality at a number of stations, including the status and trends in TSS levels. Although data are limited, these reports illustrate how DEQ will continue to work with ODA to focus work on agricultural lands to reduce sedimentation and mercury loading. DEQ will take a similar approach with both federal and non-federal forest lands.

Among both point and nonpoint sources, there is variation as to maturity of programs focused on mercury minimization measures. DEQ anticipates that entities with longer experience in implementing measures targeting mercury, particularly erosion and runoff controls, will continue to achieve modest reductions using strategies and techniques that have evolved over time.

Oregon Department of Environmental Quality 138 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

DEQ expects that entities that have not yet begun implementing mercury minimization measures can learn from practices employed by entities with more mature programs. In addition, DEQ expects that entities employing these techniques and strategies for the first time have greater reduction potential. Together, optimized mercury reduction actions applied broadly across all sources is anticipated to achieve the aggregated sector-specific allocations over time. 14.1.5 Take action on failure to implement Following up on reviews to track progress of implementation plans, DEQ will take appropriate action if the DMAs or responsible persons fail to develop or effectively implement their implementation plan or fulfill milestones. DEQ’s actions can take two tracks, enforcement or engagement in voluntary initiatives. DEQ uses both, as appropriate within the process, to achieve optimal pollutant reductions. In some cases DEQ can assist in facilitating the availability of incentives for meeting voluntary initiatives or providing education. DEQ will also take enforcement actions where necessary based on authorities listed in Section 13.8, or raise the issue to the EQC as provided in OAR 340-042-0080. 14.1.6 Track water quality status and trends As noted above in Section 13.6, DEQ is tracking water quality status and trends concurrently as management strategies are implemented. DEQ is relying on a system of interconnected evaluations, which include DMAs meeting measurable objectives, effectiveness demonstration of mercury management strategies, accountability of implementation, discharge monitoring and instream monitoring. Together, these data and evaluations will allow refinement of focus on specific geographic areas or discharges and appropriate implementation of adaptive management actions to attain, over time, the objectives of the TMDL. In partnership with EPA, DEQ is currently developing an Assessment and Monitoring Strategy to Support Implementation of Mercury Total Maximum Daily Loads for the Willamette Basin. Intended to be a living document, this plan will serve as the overarching structure to tie together the information gained from the other evaluations during implementation of the TMDL by the 171 DMAs and responsible persons.

Tracking of water quality status and trends will include DEQ tracking and reporting on: • TMDL implementation plan submittals, reviews, and approvals • DMA, responsible person and permittee implementation of management actions • Instream compliance points for allocations, in conjunction with revisiting the watershed modeling • Annual and other increment reporting from DMAs, responsible persons and permittees • Five year reviews of implementation and evaluation of the TMDL and WQMP 14.2 Dominance of atmospheric deposition of mercury As discussed in the TMDL Technical Support Document and preceding sections of this draft TMDL, atmospheric deposition of mercury onto the Oregon landscape is the dominant source of mercury reaching Willamette Basin streams. While these deposited air emissions originate as a mix of global, national, regional and local sources, the largest portion is derived from historical deposition of global anthropogenic mercury emissions (TetraTech, 2019a). Further, the current air emissions sources originating within Oregon are small relative to the total mercury budget of the basin. Air emissions from local sources are being addressed by existing programs and

Oregon Department of Environmental Quality 139 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

mercury loads from all permitted point source discharges combined are conservatively estimated to be less than five percent of the total mercury load or approximately 18 g/day or 6.61 kg/yr. As such there is limited overall potential for reducing mercury loads within Willamette Basin streams through further reductions of air emissions and wastewater discharge point sources. Despite distant origins of the dominant sources of mercury, once on the landscape in the Willamette Basin, the greatest potential for reductions of mercury delivered to streams is through enhancing controls on nonpoint source land use activities that have the potential to result in erosion and surface runoff. DEQ’s approach prioritizes focus on controls for erosion and surface runoff from both point and nonpoint sources to optimize mercury reductions into waterways.

In alignment with EPA guidance relevant to the Willamette Basin situation where mercury loadings are predominantly from air deposition (EPA 2008, 2010), DEQ opted to allocate aggregated nonpoint source loads and point source wasteloads using the proportionality approach. These approaches also follow precedents affirmed in EPA-approved mercury TMDLs in 21 other states. These allocations include portions of natural and anthropogenic background sources that are outside of the reasonable control of designated management agencies and responsible parties. 14.3 Conclusions DEQ’s implementation approach is multi-faceted and requires many targeted management practices across the entire basin to reduce anthropogenic mercury, regardless of source origination. This is a reasonable approach that recognizes the inherent uncertainty in global atmospheric deposition reduction trends, on-going inputs from historical sources of mercury still available to be delivered to streams and long lag times until positive responses occur in streams and fish.

Because the depositional sources are mixed and the management practices that can be employed are distributed over a wide area and among many DMAs and responsible persons, there is uncertainty about reductions in mercury loading. DEQ’s draft WQMP addresses this uncertainty by including an extensive monitoring, reporting, and adaptive component that is designed to match the accountability framework used by EPA in its Chesapeake Bay TMDL (2010).

The examples of effective actions employed since issuance of the 2006 TMDL (presented in Section 14.1.4 above), demonstrate that effective mercury management practices exist that can and will be employed, for both nonpoint and point source activities, to achieve the load and wasteload allocations contained in this draft TMDL.

The rationale described in this document stems from a more robust evaluation using significantly more data, captures additional urban areas not previously regulated, implements an accountability framework (including the Assessment and Monitoring Strategy to Support Implementation of Mercury Total Maximum Daily Loads for the Willamette Basin) and provides opportunities for adaptive management to maximize mercury reductions. Together this approach provides reasonable assurance to meet state and federal requirements and attain the goals of the TMDL.

Oregon Department of Environmental Quality 140 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 15. References

Amos, H. M., Jacob, D. J., Streets, D. G., & Sunderland, E. M. (2013). Legacy impacts of all- time anthropogenic emissions on the global mercury cycle. Global Biogeochemical Cycles, 27, 410-421. doi:10.1002/gbc.20040 Balogh, S., Nollet, Y., & Offerman, H. (2005). A comparison of total mercury and methylmercury export from various Minnesota watersheds. Science of the Total Environment, 340, 261- 270. Benoit, J., Gilmour, C., Heyes, A., Mason, R., & Miller, C. (2002). Geochemical and biological controls over methylmercury production and degradation in aquatic ecosystems. In Biogeochemistry of Environmentally Important Trace Elements (Vol. 835, pp. 262-297). doi:10.1021/bk-2003-0835.ch19 California Environmental Protection Agency. (2010). Sacramento -- San Joaquin Delta Estuary TMDL for Methylmercury Staff Report. Regional Water Control Board, Central Valley Region. California Environmental Protection Agency. (2017). Final part 2 of the Water Quality Control Plan for inland surface waters, enclosed bays, and estuaries of California -- Tribal and subsistence fishing beneficial uses and mercury provisions. Regional Water Control Board, Central Valley Region. Chetelat, J., Amyot, M., Arp, P., Blais, J., Depew, D., Emmerton, C., . . . van der Velden, S. (2015). Mercury in freshwater ecosystems of the Canadian Arctic: Recent advances on its cycling and fate. Science of the Total Environment, 509, 41-66. Christensen, G., Wymore, A., King, A., Podar, M., Hurt, R., & Santillan, E. (2016). Development and Validation of Broad-Range Qualitative and Clade-Specific Quantitative Molecular Probes for Assessing Mercury Methylation in the Environment. Applied and Environmental Microbiology, 82, 6068-6078. Colorado Department of Public Health and Environment. (2003). TMDL for Mercury in McPhee and Narraguinnep Reservoirs, Colorado, Phase I. Conroy, M. J., & Peterson, J. T. (2013). Decision Making in Natural Resource Management: A Structured, Adaptive Approach. Hoboken: Wiley-Blackwell. Dietrich, L. (2019, February 28). Personal Communication. Wisconsin Department of Natural Resources. Domagalski, J., Majewski, M., Alpers, C., Eckley, C., Eagles-Smith, C., Schenk, L., & Wherry, S. (2016). Comparison of mercury mass loading in streams to atmospheric deposition in watersheds of Western North America: Evidence for non-atmospheric mercury sources. Science of the Total Environment, 568, 638-650. doi:10.1016/j.scitotenv.2016.02.112 Driscoll, C., Han, Y., Chen, C., Evers, D., Lambert, K., & Holsen, T. (2007). Mercury contamination in forest and freshwater ecosystems in the Northeastern United States. Bioscience, 57, 17-28. Eagles-Smith, C., Ackerman, J., Willacker, J., Tate, M., Lutz, M., & Fleck, J. (2016a). Spatial and temporal patterns of mercury concentrations in freshwater fish across Western United States and Canada. Science of the Total Environment, 568, 1171-1184. Eagles-Smith, C., Silbergeld, E., Basu, N., Bustamante, P., Diaz-Barriga, F., & Hopkins, W. (2018). Modulators of mercury risk to wildlife and humans in the context of rapid global change. Ambio, 47, 170-197. Eagles-Smith, C., Wiener, J., Eckley, C., Willacker, J., Evers, D., & Marvin-DiPasquale, M. (2016b). Mercury in western North America: A synthesis of environmental contamination, fluxes, bioaccumulation, and risk to fish and wildlife. Science of the Total Environment, 568, 1213-1226.

Oregon Department of Environmental Quality 141 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Eckley, C., & Branfireun, B. (2009). Mercury mobilization in urban stormwater runoff. Science of the Total Environment, 403, 164-177. Eckley, C., & Branfireun, B. (2009). Simulated rain events on an urban roadway to understand the dynamics of mercury mobilization in stormwater runoff. Water research, 43(15), 3635-3646. Eckley, C., Eagles-Smith, C., Tate, M., Kowalksi, B., Danehy, R., Johnson, S., & Krabbenhoft, D. (2018). Stream mercury export in response to contemporary timber harvesting methods (Pacific coastal mountains, Oregon, USA). Environ. Sci. Technol, 52. Eckley, C., Luxton, T., Goetz, J., & McKernan, J. (2017). Water-level fluctuations influence sediment porewater chemistry and methylmercury production in a flood-control reservoir. Environmental Pollution, 222, 32-41. Eckley, C., Tate, M., Lin, C.-J., Gustin, M., Dent, S., & Eagles-Smith, C. (2016). Surface-air mercury fluxes across Western North America: A synthesis of spatial trends and controlling variables. Science of the Total Environment, 568, 651-665. Fitzgerald, W., Engstrom, D., Mason, R., & Nater, E. (1998). The case for atmospheric mercury contamination in remote areas. Environmental Science and Technology, 32, 1-7. Fleck, J., Alpers, C., Marvin-DiPasquale, M., Hothem, R., Wright, S., Ellett, K., . . . May, J. (2010). The effects of sediment and mercury mobilization in the South Yuba River and Humbug Creek Confluence Area, Nevada County, California. Concentrations, speciation, and environmental fate -- Part 1: Field characterization, US Geological Survey Open-File Report 2010-1325A. Florida Department of Environmental Protection. (2013). Mercury TMDL for the State of Florida. FTN Associates Ltd. (2007). Mercury TMDL for Spring Lake in Yell County, Arkansas. Prepared for EPA Region 6. Graham, A., Aiken, G., & Gilmour, C. (2012). Dissolved organic matter enhances microbial mercury methylation under sulfidic conditions. Environmental Science & Technology, 46, 2715-2723. Gray, J., Hines, M., Krabbenhoft, D., & Thoms, B. (2012). Methylation of Hg downstream from the Bonanza Hg Mine, Oregon. Applied Geochemistry, 27, 106-114. HAdjuk, G. (2017). Introduction to linear mixed models. Retrieved from https://ourcodingclub.github.io/2017/03/15/mixed-models.html Hall, B., Bodaly, R., Fudge, R., Rudd, J., & Rosenberg, D. (1997). Food as the dominant pathway of methylmercury uptake by fish. Water, Air and Soil Pollution, 100, 13-24. Henny, C., Kaiser, J., Packard, H., Grove, R., & Taft, M. (2005). Assessing mercury exposure and effects to American Dippers in headwaters streams near mining sites. Ecotoxicology, 14, 709-725. Hillwig, R. (2019, January 17). Personal communication via phone; on OHA's current method for determining fish advisories. 2019: Oregon Health Authority -- Public Health Division. Hope, B. (2003). A basin-specific aquatic food web biomagnification model for estimateion of mercury target levels. Environmental Toxicology and Chemistry, 22, 2525-2537. Hsu-Kim, H., Eckley, C., Ach`a, D., Feng, X., Gilmour, C., Jonsson, S., & Scott, C. (2018). Challenges and opportunities for managing aquatic mercury pollution in altered landscapes. Ambio, 47(2), 141-169. Hsu-Kim, H., Kucharzyk, K., Zhang, T., & Deshusses, M. (2013). Mechanisms regulating mercury bioavailability for methylating microorganisms in the aquatic environment: a critical review. Environmental Science & Technology, 47, 2441-2456. Humphreys, R. (2005). Losses and recovery during a suction dredge test in the south fork of the American River. Staff Report, State Water Resources Control Board, Division of Water Quality. Hygelund, B., Ambers, R., & Ambers, C. (2001). Tracing the source of mercury contamination in Dorena Lake Watershed, Western Oregon. Environmental Geology, 40, 853-859. doi:10.1007/s002540100245

Oregon Department of Environmental Quality 142 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Idaho Department of Environmental Quality. (2009). Jordan Creek Subbasins Assessment and TMDL. Jaeger, A., Plantinga, A., Langpap, C., Bigelow, D., & Moore, K. (2017). Water, economics, and climate change in the Willamette Basin, Oregon. Oregon State University Extension Service EM 1957. Retrieved from https://catalog.extension.oregonstate.edu/sites/catalog/files/project/pdf/em9157.pdf Kammerer, J. (May 1990). Water fact sheet -- Largest Rivers in the United States. U.S. Geological Survey, Department of Interior . Keyes, A., Lambert, K., Burtraw, D., Buonocore, J., Levy, J., & Driscoll, C. (2018). Carbon Standards Examined: A comparison of at-the-source and beyond-the-source power plant carbon standards. Proposed Affordable Clean Energy Rule: Emissions Consequences Fact Sheet. Resources for the Future. . Kocman, D., Wilson, S., Amos, H., Telmer, K., Steenhuisen, F., & Sunderland, E. (2017). Toward an assessment of the global inventory of present-day mercury releases to freshwater environments. International Journal of Environmental Research and Public Health, 14. Limno Tech. (2018). Michigan Statewide Mercury TMDL. Prepared for Michigan Department of Environmental Quality and U.S. Environmental Protection agency, Region 5, Limno Tech under subcontract to Battelle. Lindberg, S., Bullock, R., Ebinghaus, R., Engstrom, D., Feng, X., & Fitzgerald, W. (2007). A synthesis of progress and uncertainties in attributing the sources of mercury in deposition. Ambio, 36, 19-32. Marvin-DePasquale, M., Lutz, M., Brigham, M., Krabbenhoft, D., Aiken, G., Orem, W., & Hall, B. (2009). Mercury cycling in stream ecosystems. 2. Benthic mehtylmercury production and bed sediment pore water partitioning. Environ Sci Technol, 43, 2726-2732. Marvin-DiPasquale, M., Agee, J., Kakouros, E., Kieu, L., Fleck, J., & Alpers, C. (2011). The effects of sediment and mercury mobilization in the South Yuba River and Humbug Creek Confluence Area, Nevada County, California. Concentrations, speciation and environmental fate -- Part 2: Laboratory Experiments, US Geolgoical Survey Open-File Report 2010-1325B. Maryland Department of the Environment. (2002). TMDL of Mercury for Pretty Boy Reservoir, Baltimore County, Maryland. Maryland Department of the Environment. (2014). Mercury Stormwater Implementation Plan and NPDES MS4 Guidance. Minnesota Pollution Control Agency. (2007). Minnesota Statewide Mercury TMDL. Montana Department of Environmental Quality. (2013). Lake Helena planning area metals TMDL addendum. Munthe, J., Bodaly, R., Branfireun, B., Driscoll, C., Gilmour, C., & Harris, R. (2007). Recovery of mercury-contaminated fisheries. Ambio, 36, 33-44. New England Interstate Water Pollution Control Commission. (2007). Northeast Regional Mercury TMDL. Vermont Department of Environmental Conservation, New York Department of Environmental Conservation, Connecticut Department of Environmental Protection, Maine Department of Environmental Protection, Massachusetts Department of Environmental Protection, New Hampshire Department of Environmental Services, Rhode Island Department of Environmental Management. New Jersey Department of Environmental Protection. (2009). New Jersey Mercury Reduction Action Plan. Mercury Work Group. North Carolina Department of the Environment and Natural Resources. (2012). Mercury post- TMDL permitting strategy. Division of Water Quality. North Carolina Department of the Environment and Natural Resources. (2012). North Carolina Mercury TMDL. Division of Water Quality.

Oregon Department of Environmental Quality 143 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

North Carolina Department of the Environment and Natural Resources. (2012). North Carolina's mercury reduction options for nonpoint sources. Division of Water Quality. Northwest Environmental Advocates v. United States Environmental Protection Agency, 3:12- cv-01751-AC (US District Court for the District of Oregon, Document 133; October 12, 2016). Northwest Environmental Advocates v. United States Environmental Protection Agency, 3:12- cv-01751-AC (US District Court for the District of Oregon, Document 133; September 12, 2016). Northwest Environmental Advocates v. United States Environmental Protection Agency, 3:12- cv-0751-AC (US District Court for the District of Oregon, Document 149; April 11, 2017). Obrist, D., Pearson, C., Webster, J., Kane, T., Lin, C., & Aiken, G. (2016). A synthesis of terrestrial mercury in the western United States: spatial distribution defined by land cover and plant productivity. Science of the Total Environment, 568, 522-535. Oregon Department of Environmental Quality. (2006). Willamette Basin Total Maximum Daily Load. Retrieved from https://www.oregon.gov/deq/wq/tmdls/Pages/TMDLs-Willamette- Basin.aspx Oregon Department of Environmental Quality. (2017). Willamette Basin Mercury Total Maximum Daily Load Advisory Committee Charter. Retrieved from https://www.oregon.gov/deq/FilterDocs/WBACcharter.pdf Oregon Department of Environmental Quality. (2018). DEQ Integrated Toxics Reduction Strategy -- 2018 Update. Retrieved from https://www.oregon.gov/deq/FilterDocs/ToxicsStrategy.pdf Oregon Department of Environmental Quality. (2019a). Willamette River instream surrogate TSS-THg analysis. Oregon Department of Environmental Quality. (2019b). Draft Mercury Multipe Discharger Variance for the Willamette Basin and amendments to Oregon Variance Rule. Oregon Health Authority. (2016). Oregon Statewide Bass Fish Consumption Advisory due to Mercury Contamination. Technical Report, Public Health Division, Portland, Oregon. Oregon Health Authority. (2019). Fish advisories and consumption guidelines webpage. Retrieved February 5, 2019, from https://www.oregon.gov/oha/PH/HEALTHYENVIRONMENTS/RECREATION/FISHCON SUMPTION/Pages/fishadvisories.aspx#fish Schmidt, M. (2018). Willamette River Basin Mercury Data Summary (Draft). Memorandum from Michelle Schmidt to Jayshika Ramrakha, Leigh Woodruff, Alan Henning (USEPA), and Paula Calvert (ODEQ). March 6, 2018. Research Triangle Park, NC. Schroeder, W., & Munthe, J. (1998). Atmospheric mercury -- An overview. Atmospheric Environment, 30, 809-822. Seigneur, C. K. (2004). Global source attribution for mercury deposition in the United States. Environmental Science and Technology, 38, 555-569. Shanley, J., Mast, M., Campbell, D., Aiken, G., Krabbenhoft, D., & Hunt, R. (2008). Comparison of total mercury and methylmercury cycling at five sites using the small watershed approach. Environmental Pollution, 154, 143-154. Skyllberg, U., Xia, K., Bloom, P., Nater, E., & Bleam, W. (2000). Binding of mercury(II) to reduced sulfur in soil organic matter along upland-peat soil transects. Journal of Environmental Quality, 29, 855-865. Smith, D., Cannon, W., Woodruff, L., Solano, F., & Ellefsen, K. (2014). Geochemical and mineralogical maps for soil of the conterminous United States. (e. Survey USG, Ed.) U.S. Geological Survey Open-File Report 2014-1082, p. 386. South Dakota Department of Environmental and Natural Resources. (2015). South Dakota Mercury TMDL. Strode, S., Jaegle, L., Jaffe, D., Swartzendruber, P., Selin, N., & Holmes, C. (2008). Trans- Pacific transport of mercury. Journal of Geophysical Research-Atmospheres, 113.

Oregon Department of Environmental Quality 144 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Tetra Tech. (2019b). Memorandum on Willamette River Basin Mercury TMDL-- Tualatin Sediment Reduction Scenarios. TetraTech. (2019a). Mercury TMDL development for the Willamette River Basin (Oregon). Public Review Draft. Technical Support Document. Tobias, R., & Wasley, D. (2013). 2012 Annual Post Removal Action Monitoring Report Champion Mine -- Umpqua National Forest, Lane County, Oregon. Technical Memorandum. Trip, L., & Allan, R. (2000). Sources, trends, implications and remediation of mercury contamination of lakes in remote areas of Canada. Water Science and Technology, 42, 171-176. U.S. Environmental Protection Agency. (1991). Guidance for Water Quality-Based Decisions: The TMDL Process. EPA 440/4-01-001, Office of Water, Washington, DC. U.S. Environmental Protection Agency. (2001). Water Quality Criterion for the Protection of Human Health. Methyl mercury. U.S. Environmental Protection Agency. (2002). Memorandum on establishing Total Maximum Daily Load (TMDL) Wasteload Allocations (WLAs)for stormwater sources and NPDES permit requirements based on those WLAs. Office of Water, Washington, DC. U.S. Environmental Protection Agency. (2002). Method 1631, Revision E: Mercury in Water by Oxidation, Purge and Trap, and Cold Vapor Atomic Fluorescence Spectrometry. Retrieved from https://www.epa.gov/sites/production/files/2015- 08/documents/method_1631e_2002.pdf U.S. Environmental Protection Agency. (2002). TMDL development for total mercury in fish tissue residue in the Middle and Lower Savannah River Watershed. For segments Clarks Hill Lake Dam to Stevens Creek Dam, Stevens Creek Dam to US Highway 78/278, US Highway 78/278 to Johnsons Landing, Johnsons Landing to Brier Creek, Brier Creek to teh Tide Gate, USEPA, Region4. U.S. Environmental Protection Agency. (2007). An Approach Using Load Duration Curves in the Development of TMDLs. Technical Document, Office of Wetlands, Oceans and Watersheds, Watershed Branch, Washington D.C. U.S. Environmental Protection Agency. (2008a). TMDLs where mercury loadings are predominantly from air depositions. Office of Water, Washington, DC. U.S. Environmental Protection Agency. (2008b). Draft TMDLs to Stormwater Permits Handbook. Office of Wetlands, Oceans and Watersheds; Office of Wastewater Management; Region 5 Water Division, Washington, DC.; Chicago, IL. U.S. Environmental Protection Agency. (2010a). Guidance for implementing the January 2001 Methylmercury Water Quality Criterion. EPA 823-R-10-001, Office of Water, Washington, DC. U.S. Environmental Protection Agency. (2010b). MS4 permit improvement guide. EPA 833-R- 10-001, Office of Wastewater Management, Washington, DC. U.S. Environmental Protection Agency. (2011). National Scale Assessment of Mercury Risk to Populations with High Consumption of Self-Caught Freshwater Fish in Support of the Appropriate and Necessary Finding for Coal and Oil-Fired Electric Generating Units. Revised Technical Support Document, Office of Air Quality Planning and Standards. U.S. Environmental Protection Agency. (2014). Memorandium on revisions to the November 22, 2002 memorandum "Establishing Total Maximum Daily Load (TMDL) wasteload allocations (WLAs) for stormwater sources and NPDES permit requirements based on those WLAs. Office of Water, Washington, DC. U.S. Environmental Protection Agency. (2015). Helpful practices for addressing point sources and implementing TMDLs in NPDES permits. USEPA, Region 9. U.S. Environmental Protection Agency. (2016). Compendium of MS4 permitting approaches. EPA-810-U-16-001, Office of Wastewater Management.

Oregon Department of Environmental Quality 145 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Ullrich, S., Tanton, T., & Abdrashitova, S. (2001). Mercury in the aquatic environment: A review of factors affecting methylation. Critical Reviews in Environmental Science and Technology, 31, 241-293. United Nations Environment Programme. (2019). Global Mercury Assessment 2018. Geneva, Switzerland: UN Environment Programme, Chemicals and Health Branch. US Army Corps of Engineers. (2017). Willamette Basin review feasibility study: Draft integrated feasibility report and environmental assessment. US Army Corps of Engineers, Portland District. Retrieved from https://usace.contentdm.oclc.org/digital/collection/p16021coll7/id/8219 Virginia Department of Environmental Quality. (2009). TMDL development for mercury in South River, South Fork Shenandoah River and Shenandoah River, VA. Warner, K., Roden, E., & Bonzongo, J. (2003). Microbial mercury transformation in anoxic freshwater sediments under iron-reducing and other electron-accepting conditions. Environmental Science & Technology, 37, 2159-2165. Wentz, D., Bonn, B., Carpenter, K., Hinklel, S., Janet, M., Rinella, F., . . . Bencala, K. (1998). Water quality in the Willamette Basin, Oregon, 1991-1995. US Geological Survey Circular 1161. Retrieved from htt;s://pubs.usgs.gov/circ/circ1161/circ1161.pdf Willacker, J., Eagles-Smith, C., Lutz, M., Tate, M., Lepak, J., & Ackerman, J. (2016). Reservoirs and water management influence fish mercury concentrations in the western United States and Canada. Science of the Total Environment, 568, 739-748. Wright, L., Zhang, L., & Marsik, F. (2016). Overview of mercury dry deposition, litterfall, and throughfall studies. Atmospheric Chemistry and Physics, 16, 13399-13416.

Oregon Department of Environmental Quality 146 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 Appendix A. Technical Support Document

The Technical Support Document is available at: https://www.oregon.gov/deq/wq/tmdls/Pages/willhgtmdlac2018.aspx

Oregon Department of Environmental Quality 147 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 Appendix B. Calculation of allocations

The allocations for the TMDL were calculated as percent reductions and compared to the load capacity. These calculations and comparisons were automated and done in a spreadsheet, located on DEQ’s website: https://www.oregon.gov/deq/wq/tmdls/Pages/willhgtmdlac2018.aspx. The existing loads for the mercury sources obtained from the Mass Balance Model were entered into the spreadsheet and then criteria were set for the optimization routines. The criteria pertained to the size of the reductions in a source that were being tested. The source categories used in the spreadsheet are listed in Table B-1. The percent reductions listed in the second column of Table B-2 were multiplied by the loads for each of the source categories. The third column is the result that the reduction of atmospheric deposition load will result in a reduction in the load in surface runoff in addition to a reduction in loads directly from surface runoff. The fourth and fifth columns of Table B-2 are the limits in the reductions set for the optimization. The optimization was setup so the smallest reduction within the range would be used in the load calculation. The optimization was run until the calculated TMDL load was less than or equal to the load capacity as shown in Table B-3.

Table B-1. Source category names and descriptions used in spreadsheet for calculating the allocations.

Source Category Description Runoff of atmospheric THg in surface runoff from wet and dry atmospheric deposition to pervious deposition and impervious surfaces. Atmospheric deposition THg from wet and dry atmospheric deposition direct to water surfaces. direct to water Dissolved THg associated with subsurface flows (shallow interflow and Groundwater resurfacing groundwater). Sediment Particulate-associated THg from sediment erosion and transport. THg associated with stormwater in permitted Phase I & II MS4 areas (cities, counties, & ODOT). Includes THg from runoff of atmospheric deposition, MS4 dissolved THg in shallow subsurface interflow, and particulate-associated THg from sediment erosion. Limited to developed low, medium, and high density land in MS4 areas. THg associated with stormwater from urban Designated Management Areas (DMAs). Includes THg from runoff of atmospheric deposition, dissolved THg Urban DMAs in shallow subsurface interflow, and particulate-associated THg from sediment erosion. Limited to developed low, medium, and high density land in urban DMA areas. Legacy THg from contaminated mine tailings and furnace areas at the Black Mines Butte Mine and the Bohemia Mining District.

THg associated with discharges from major POTWs and minor domestic POTWs WWTPs that hold NPDES permits. THg associated with discharges from industrial facilities that hold NPDES Industrial dischargers permits.

Oregon Department of Environmental Quality 148 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Table B-2. Allocation spreadsheet table. Combined with Opt Atmospheric Lower Upper Category Reduction Deposition Limit Limit Reduction Runoff of atmospheric deposition 88% (agriculture) 99% 80% 88% Runoff of atmospheric deposition (forest) 88% 99% 80% 88% Runoff of atmospheric deposition (shrub) 88% 99% 80% 88% Runoff of atmospheric deposition 88% (developed) 99% 80% 88% Runoff of atmospheric deposition (other) 88% 99% 80% 88% Atmospheric deposition direct to water 11% 0% 11% Groundwater (agriculture) 32% 0% 88% Groundwater (forest) 88% 0% 88% Groundwater (shrub) 77% 0% 88% Groundwater (developed) 0% 0% 88% Groundwater (other) 88% 0% 88% Sediment (agriculture) 88% 75% 88% Sediment (forest) 88% 0% 88% Sediment (shrub) 88% 80% 88% Sediment (developed) 88% 75% 88% Sediment (other) 88% 75% 88% MS4 75% 50% 75% Urban DMAs 75% 50% 75% Mines 95% 90% 95% POTWs 10% 10% 75% Industrial dischargers 10% 10% 75%

Table B-3. Comparison of Load Capacity to TMDL Loads Calculated for the Reductions. Category Load Load Capacity (g/day) 42.17 TMDL for reductions (g/day) 42.17

Oregon Department of Environmental Quality 149 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 Appendix C. Variance justification excerpts

Water Quality Based Effluent Limits for mercury are not technologically achievable

There are no technology-based effluent limits or effluent limitations guidelines for mercury. Therefore, NPDES permit limits for mercury are based on the water quality criterion. Because total mercury levels in the Willamette River basin exceed the water concentration needed to meet the fish tissue-based methylmercury criterion, dischargers would be required to achieve an effluent concentration equal to the water concentration target of 0.14 ng/L before the effluent is discharged to the receiving water. Current treatment technology can reliably attain concentrations less than 20 ng/L. Treatment achieving these levels is typically through the removal of solids which have mercury adsorbed to them. Mercury removal is an ancillary benefit of wastewater treatment and effluent concentrations vary significantly, even when influent concentrations are similar. Moreover, any removed mercury from treatment is likely to end up in biosolids, which are then disposed of through land application or to landfills, where the mercury can re-enter the environment. DEQ also examined other treatment technologies and determined there are currently no feasible treatment technologies that could feasibly reduce mercury levels enough to achieve an effluent concentration of 0.14 ng/L.

Mercury Levels Currently Achieved by Secondary and Advanced Wastewater Treatment Plants

The information in this section demonstrates that, in general, current wastewater treatment technology, while removing 90% or more of mercury from influent, consistently achieves average mercury concentrations ranging from 1-20 ng/L. However mercury removal is an ancillary benefit of treatment, mercury concentrations are so small, and mercury can enter into a collection system in unexpected ways. As a result, effluent concentrations vary significantly, even in effluent from one discharger and under similar conditions.

In 2005, California performed a study looking at methylmercury removal from NPDES permitted dischargers in the Sacramento River Delta (California EPA, 2010). California required dischargers to collect and report on methylmercury influent and effluent data over twelve months in 2004 and 2005. Ofthese facilities, 49 also reported total mercury effluent data. The median of the average annual total mercury effluent concentrations from these facilities was 6.6 ng/L and ranged from 0.8-21.5 ng/L.

DEQ also compiled and analyzed mercury levels from 2016 data provided by 19 municipal dischargers in Oregon that have pre-treatment programs. The median average annual total mercury effluent concentration was 2.1 ng/L and ranged from 1.1 to 8.3 ng/L.

This information, along with Wisconsin data presented in Section 2, indicate that current treatment achieves a mercury concentrations as low as 1 ng/L and sometimes much higher. However, no treatment plants using current treatment technology are able to achieve mercury levels as low as 0.14 ng/L, which would be required without a variance.

Oregon Department of Environmental Quality 150 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Mercury Levels Achieved by Other Treatment Technologies

In reviewing the ability of other available wastewater treatment technologies to remove mercury, DEQ could not find any pilot or full-scale treatment systems that would be able to achieve the water concentration target of 0.14 ng/L.

Because there is a lack of full-scale installations consistently producing effluent mercury concentrations less than those found in current treatment, it is difficult to predict whether it is possible to consistently achieve these concentrations on a long-term, large-scale basis. A 1997 study in Ohio concluded that the ability of the added controls to meet the standard was not known (Ohio EPA, 1997). The Ohio mercury criterion for aquatic life is 1.3 ng/L. In Oregon, the WQBEL needed to meet the human health criterion for methylmercury is 0.14 ng/L, an order of magnitude lower than the Ohio and Michigan standards. If the ability of the controls to meet 1.3 ng/L is not known, it is reasonable to conclude that there is no feasible technology that can meet 0.14 ng/L.

This conclusion is consistent with a review conducted by HDR in 2013 for the Association of Washington Businesses (HDR, 2013). The HDR study examined the potential performance of adding reverse osmosis or granular activated carbon to a tertiary microfiltration process and hypothesized that such a treatment system might be able to remove mercury to a concentration of 0.12 to 1.2 ng/L. However, the study provided no data from any test or operational system. Such treatment systems have not been employed on a bench or pilot scale, or at a wastewater treatment plant scale to DEQ’s knowledge.

Membrane filtration technology, such as reverse osmosis, uses a significant amount of electricity, creating a substantial carbon footprint, and requires disposal of waste brine. According to a life cycle assessment performed for the Berlin-Ruhleben secondary wastewater treatment plant (63 MGD), the operational energy use of polymer ultrafiltration or ceramic microfiltration membranes would be 0.33 watt×hour/gal. This would represent approximately a 9 percent increase in that plant's existing global warming potential without taking into account the additional global warming potential that would be contributed by the infrastructure, chemicals for maintenance and any necessary coagulant, and the transport of waste sludge for disposal. Of the different types of membrane filtration, reverse osmosis also has the large disadvantage of necessitating disposal of the concentrate stream, which can amount to approximately 5 to 20 percent of the influent.

A 2007 EPA report regarding mercury treatment notes that there are technologies, such as precipitation, filtration or other physical/chemical treatments (see Table C-1) that remove more mercury than secondary or advanced wastewater treatment plants. However, these have been employed in industrial settings where influent concentrations were an order of magnitude higher than influent concentrations at municipal wastewater treatment facilities (U.S. EPA, 2007). The effluent concentrations at many of these industrial applications were similar to the influent concentrations at municipal treatment facilities. Moreover, the information provided in the EPA report did not indicate flow volumes, so it is difficult to translate these studies to typically larger municipal wastewater treatment plant volumes.

In another study, an oil refinery evaluated various treatment technologies for wastewater with low (10 ng/L) mercury levels to determine the extent to which mercury concentrations could be further reduced using conventional treatment. Bench scale tests of various adsorbent techniques showed that they could remove mercury to as low as less than 0.08 ng/L of total mercury (Urgun-Demirtas, et al., 2013). Ultra- and micro-filtration bench tests also reduced

Oregon Department of Environmental Quality 151 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 mercury to less than 1 ng/L, although not as much as adsorption. However, such techniques have not been shown to work at the higher volume in municipal treatment (HDR, 2013).

Table C-1 shows the results from treatment technologies that have been tested for water supply treatment or industrial wastewater treatment. Table C-2 summarizes mercury concentrations achieved from various technologies. As shown in these tables, no technology has consistently reached mercury concentrations less than that achieved by activated sludge (secondary treatment) or activated sludge with nutrient removal or tertiary filtration (advanced treatment) at flow volumes typically seen at large municipal WWTPs (>1 MGD).

Table C-1. Potential treatment technologies considered for mercury treatment Influent total Average effluent Type of mercury total mercury Percent Study treatment concentration concentration removal technology (ng/L) (ng/L) US EPA Precipitation 400-9,600,000 25-21,400 42-99.9% Full scale for (2007) (Chelator) groundwater and wastewater treatment; not tested for municipal wastewater or industrial processes in Willamette Basin EPA (2007) Adsorption/ 3,300-2,500,000 300-1,000 99-99.8%% Full scale Granular Activated Carbon HDR Study Tertiary 0.12-1.2 >99% Not (2013) Microfiltration/ hypothetically demonstrated Reverse at WWTP scale Osmosis or Granular Activated Carbon Urgun- Precipitation 10 ng/L 3.1 ng/L (before 56.5% Bench scale Demirtas, filtration) before testing et al. 0.17 ng/L (after filtration (2013) filtration) Urgun- Adsorption 10 ng/L <0.08 ng/L – 0.72 92.8% - Bench scale Demirtas, ng/L (lowest 99.2% testing et al. achieved) (2013) Urgun- Filtration 10 ng/L 0.26 – 0.34 ng/L 65 – 97% Bench scale Demirtas, (lowest achieved) depending testing et al. on (2013) pressure Hollerman, Adsorption 739-1447 ng/L ~25-340 ng/L n/a Low volume et al. (1999)

Oregon Department of Environmental Quality 152 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Table C-2. Treatment capability of mercury technologies Volume Range of Treatment Technology Treatment Ability Known Uses Current Treatment (Secondary, Advanced Up to 25 MGD 1-20 ng/L Secondary, Tertiary) Membrane Filtration Low volume Bench scale to 0.26 ng/L Ion Exchange 0.015 MGD 1 ng/L (5-50 GPM) Precipitation and filtration Low volume Bench scale to 0.17 ng/L; full scale to 25 ng/L Adsorption Low volume Bench scale to 0.08 ng/L; full scale to 25 ng/L

The Wisconsin Department of Natural Resources has tracked mercury effluent data from NPDES permittees over the past fifteen years, as permitted facilities have been implementing MMPs under the Great Lakes Initiative. The data, as indicated in the following discussion, show that MMP implementation is an effective approach to reduce mercury.

WDNR tracks mercury concentrations using various metrics. Among 52 municipal dischargers, the average 4-day 99th percentile concentration (4-day P99) decreased from 11.2 ng/L in the initial 5-year period to 3.2 ng/L in the most recent 5-year period (2014-2018). The median 4-day P99 concentration also decreased from 5.2 to 2.8 ng/L. All but three municipal systems experienced decreasing trends in average effluent concentrations and all but eight experienced decreasing 4-day P99 concentrations (Figure C-1). Moreover, whereas 13 facilities had 4-day P99 concentrations greater than 8 ng/L in their initial permit term, only one facility had concentrations greater than 8 ng/L based on the most recent data (Figure C-2), highlighting how effluent levels have decreased over time. According to WDNR staff, these facilities have achieved mercury reductions primarily, if not exclusively, through MMP implementation rather than treatment upgrades (Dietrich, 2019).

Figure C-1. Number of Wisconsin municipal wastewater treatment systems with increasing and decreasing trends in average (left) and 4-day P99 (right) concentrations. (Wisconsin DNR). Source: Wisconsin Department of Natural Resources.

Oregon Department of Environmental Quality 153 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

40 40 38 35 35 30 30 25 22 25 20 17 20 13 15 15 8 10 5 10 5 5 1 0 0 0 0-4 ng/l 4-8 ng/l 8-15 ng/l >15 ng/l 0-4 ng/l 4-8 ng/l 8-15 ng/l >15 ng/l

Figure C-2. Number of Wisconsin municipal WWTPs by 4-day P99 mercury concentrations from initial five-year period (left) to most recent five-year period (right). Source: Wisconsin Department of Natural Resources.

Data from Wisconsin industrial dischargers also indicates that MMP implementation has resulted in an overall decreasing trend in mercury concentrations at industrial facilities. Among 24 industrial NPDES permit holders, the mean 4-day P99 concentration decreased from 25.4 to 13.7 ng/L and the median 4-day P99 concentration decreased from 14.1 to 7.2 ng/L between 2004 and 2018. Eighteen of the 24 facilities had 4-day P99 concentrations in the most recent five-year period as compared to the initial period, and sixteen had decreasing average mercury concentrations (Figure C-3). Finally, while only one additional facility had a 4-day P99 concentration less than 8 ng/L from the initial five-year period to the most recent, five fewer facilities had concentrations greater than 15 ng/L (Figure C-4). The success of these dischargers in continuing to reduce mercury indicates that industrial dischargers in the Willamette Basin can achieve similar continued success, even for those that have been implementing MMPs for several years.

6 6

16 18

Decreasing Increasing Decreased 4-day P99 Increased 4-day P99

Figure C-3. Number of Wisconsin industrial wastewater treatment systems with increasing and decreasing trends in average (left) and 4-day P99 (right) concentrations. Source: Wisconsin Department of Natural Resources.

Oregon Department of Environmental Quality 154 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Figure C-4. Number of Wisconsin industrial NPDES facilities by 4-day P99 mercury concentrations from initial five-year period (left) to most recent five-year period (right). Source: Wisconsin Department of Natural Resources.

Evidence from influent and biosolids data also indicates the effectiveness of MMPs in reducing mercury. A decade of mercury influent data from 72 major NPDES wastewater treatment plants in Minnesota indicate that MMPs resulted in significant and continued reductions in mercury concentrations entering treatment systems. Between 2008 and 2017, influent total mercury concentrations decreased from an average of 180 ng/L to 70 ng/L (Figure C-5). Data from Oregon’s Rock Creek Advanced Wastewater Treatment Plant operated by Clean Water Services indicates decreasing mercury levels in biosolids, showing the effectiveness of their mercury reduction efforts over the last 20 years (Figure C-6).

10000

1000

100

10 Hg Conc. (ng/L) 1

0.1

0.01 2006 2008 2009 2010 2012 2013 2014 2016 2017 2019

Figure C-5. Influent Data from Major Wastewater Treatment Plants in Minnesota. Source: Minnesota Pollution Control Agency

Oregon Department of Environmental Quality 155 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Figure C-6. Mercury Concentrations in Biosolids, Rock Creek Wastewater Treatment Plan. Source: Clean Water Services.

References

California EPA, Regional Water Quality Control Board, Central Valley Region. 2010. Staff Report: A Review of Methylmercury and Inorganic Mercury Discharges from NPDES Facilities in California’s Central Valley.

Dietrich, L. 2019. Personal Communication. Wisconsin Department of Natural Resources. February 28, 2019.

HDR. 2013. Treatment Technology Review and Assessment. Prepared for the Association of Washington Businesses.

Ohio Environmental Protection Agency. 1997. Assessing the Economic Impacts of the Proposed Ohio EPA Water Rules on the Economy. Prepared for the Division of Surface Water by Foster Wheeler Environmental Corporation and DRI/McGraw Hill.

Urgun-Demirtas, M, P. Gillenwater, M. C. Negri, Y. Lin, S. Snyder, R. Doctor, L. Pierece and J. Alvarado. 2013. Achieving the Great Lakes Initiative Mercury Limits in Oil Refinery Effluent. Water Environment Research 85(1): 77-86

U.S. Environmental Protection Agency. 2007. Treatment Technologies for Mercury in Soil, Waste, and Water. Office of Superfund Remediation and Technology Innovation. Washington, DC. 133 pp.

Oregon Department of Environmental Quality 156 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 Appendix D. Stormwater references and resources

1. Center for Watershed Protection resources: https://www.cwp.org/mission-vision 2. Coquille TMDL Low Impact Development (LID) Implementation Tool: Guidance Document: https://www.oregon.gov/deq/FilterDocs/coqlidguidance.pdf 3. EPA Stormwater resources: https://www.epa.gov/npdes/npdes-stormwater-program 4. Low Impact Development in Western Oregon: A Practical Guide to Watershed Health: https://www.oregon.gov/deq/wq/tmdls/Pages/TMDLs-LID.aspx 5. SRF Program website: https://www.oregon.gov/deq/wq/cwsrf 6. TMDL Implementation Guidance: Guidance for Including Post Construction Elements in TMDL Implementation plans: https://www.oregon.gov/deq/FilterDocs/tmdls- 07wq004tmdlimplplan.pdf

Oregon Department of Environmental Quality 157 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 Appendix E. List of designated management agencies and responsible persons

DMA and Municipal New Responsible 2018 DEQ No. DMA NAME Land Use Stormwater/ Mercury Person Population Region Pop Status DMA? Category 1 Adair Village City Urban 860 <1K no WR 2 Albany City Urban 53,145 MS4 Phase 2 no WR 3 Amity City Urban 1,655 >1K yes WR 4 Aumsville City Urban 3,975 >1K no WR 5 Aurora City Urban 985 <1K no WR 6 Banks City Urban 1,785 MS4 Phase 1 no NWR (implemented by CWS) 7 Barlow City Urban 135 <1K no NWR 8 Beaverton City Urban 97,000 MS4 Phase 1 no NWR (implemented by CWS) 9 Brownsville City Urban 1,705 >1K no WR 10 Canby City Urban 16,800 >10 no NWR 11 Carlton City Urban 2,270 >1K yes WR 12 Coburg City Urban 1,195 >1K no WR 13 Cornelius City Urban 11,935 MS4 Phase 1 no NWR (implemented by CWS) 14 Corvallis City Urban 59,280 MS4 Phase 2 no WR 15 Cottage City Urban 10,005 >10K no WR Grove 16 Creswell City Urban 5,455 >5K no WR 17 Dallas City Urban 15,830 >10K no WR 18 Damascus City Urban no NWR 19 Dayton City Urban 2,720 >1K yes WR 20 Detroit City Urban 210 <1K no WR 21 Donald City Urban 985 <1K no WR 22 Dundee City Urban 3,230 >1K no WR 23 Durham City Urban 1,880 MS4 Phase 1 no NWR (implemented by CWS) 24 Estacada City Urban 3,400 >1K no NWR 25 Eugene City Urban 169,695 MS4 Phase 1 no WR 26 Fairview City Urban 8,990 MS4 Phase 1 no NWR 27 Falls City City Urban 955 <1K no WR 28 Forest Grove City Urban 24,125 MS4 Phase 1 no NWR (implemented by CWS) 29 Gaston City Urban 655 MS4 Phase 1 no NWR 30 Gates City Urban 485 <1K no WR

Oregon Department of Environmental Quality 158 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

DMA and Municipal New Responsible 2018 DEQ No. DMA NAME Land Use Stormwater/ Mercury Person Population Region Pop Status DMA? Category 31 Gervais City Urban 2,585 >1K no WR 32 Gladstone City Urban 11,880 MS4 Phase 1 no NWR 33 Gresham City Urban 110,505 MS4 Phase 1 no NWR 34 Halsey City Urban 935 <1K no WR 35 Happy Valley City Urban 20,945 MS4 Phase 1 no NWR 36 Harrisburg City Urban 3,660 >1K no WR 37 Hillsboro City Urban 101,920 MS4 Phase 1 no NWR (implemented by CWS) 38 Hubbard City Urban 3,305 >1K no WR 39 Idanha City Urban 140 <1K no WR 40 Independenc City Urban 9,370 >5K no WR e 41 Jefferson City Urban 3,245 >1K no WR 42 Johnson City City Urban 560 MS4 Phase 1 no NWR 43 Junction City City Urban 6,125 >5K no WR 44 Keizer City Urban 38,505 MS4 Phase 2 no WR 45 King City City Urban 3,700 MS4 Phase 1 no NWR (implemented by CWS) 46 Lafayette City Urban 4,105 >1K yes WR 47 Lake City Urban 38,215 MS4 Phase 1 no NWR Oswego 48 Lebanon City Urban 16,920 >10K no WR 49 Lowell City Urban 1,075 >1K no WR 50 Lyons City Urban 1,195 >1K no WR 51 Maywood City Urban 750 <1K no NWR Park 52 McMinnville City Urban 33,810 >10K yes WR 53 Mill City City Urban 1,865 >1K no WR 54 Millersburg City Urban 2,315 MS4 Phase 2 no WR 55 Milwaukie City Urban 20,525 MS4 Phase 1 no NWR 56 Molalla City Urban 9,625 >5K no NWR 57 Monmouth City Urban 9,890 >5K no WR 58 Monroe City Urban 625 <1K no WR 59 Mt. Angel City Urban 3,415 >1K no WR 60 Newberg City Urban 23,795 >10K no WR 61 North Plains City Urban 3,095 MS4 Phase 1 no NWR (implemented by CWS) 62 Oakridge City Urban 3,280 >1K no WR 63 Oregon City City Urban 34,860 MS4 Phase 1 no NWR 64 Philomath City Urban 4,715 MS4 Phase 2 no WR 65 Portland City Urban 648,740 MS4 Phase 1 no NWR 66 Rivergrove City Urban 505 MS4 Phase 1 no NWR 67 Salem City Urban 165,265 MS4 Phase 1 no WR 68 Sandy City Urban 10,990 >10K no NWR 69 Scappoose City Urban 7,200 >5K yes NWR

Oregon Department of Environmental Quality 159 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

DMA and Municipal New Responsible 2018 DEQ No. DMA NAME Land Use Stormwater/ Mercury Person Population Region Pop Status DMA? Category 70 Scio City Urban 920 <1K no WR 71 Scotts Mills City Urban 375 <1K no WR 72 Sheridan City Urban 6,190 >5K yes WR 73 Sherwood City Urban 19,505 MS4 Phase 1 no NWR (implemented by CWS) 74 Silverton City Urban 10,325 >10K no WR 75 Sodaville City Urban 345 <1K no WR 76 Springfield City Urban 60,865 MS4 Phase 2 no WR 77 St. Helens City Urban 13,240 >10K yes NWR 78 St. Paul City Urban 435 <1K no WR 79 Stayton City Urban 7,810 >5K no WR 80 Sublimity City Urban 2,890 >1K no WR 81 Sweet Home City Urban 9,225 >5K no WR 82 Tangent City Urban 1,250 >1K no WR 83 Tigard City Urban 52,785 MS4 Phase 1 no NWR (implemented by CWS) 84 Tualatin City Urban 27,055 MS4 Phase 1 no NWR (implemented by CWS) 85 Turner City Urban 2,085 MS4 Phase 2 no WR 86 Veneta City Urban 4,790 >1K no WR 87 Waterloo City Urban 235 <1K no WR 88 West Linn City Urban 25,830 MS4 Phase 1 yes NWR 89 Westfir City Urban 260 <1K no WR 90 Willamina City Urban 2,160 >1K yes WR 91 Wilsonville City Urban 25,250 MS4 Phase 1 no NWR 92 Wood Village City Urban 3,920 MS4 Phase 2 no NWR 93 Woodburn City Urban 24,760 >10K no WR 94 Yamhill City Urban 1,090 >1K yes WR 95 Benton County Urban 93,590 MS4 Phase 2 no WR County 96 Clackamas County Urban 419,425 MS4 Phase 1 no NWR County 97 Columbia County Urban 51,900 >10K yes NWR County 98 Lane County County Urban 375,120 MS4 Phase 2 no WR 99 Linn County County Urban 125,575 MS4 Phase 2 no WR 100 Marion County Urban 344,035 MS4 Phase 2 no WR County 101 Multnomah County Urban 813,300 MS4 Phase 1 no NWR Co 102 Polk County County Urban 82,100 MS4 Phase 2 no WR 103 Washington County Urban 606,280 MS4 Phase 1 no NWR County (implemented by CWS) 104 Yamhill County Urban 107,415 >10K yes WR County

Oregon Department of Environmental Quality 160 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

DMA and Municipal New Responsible 2018 DEQ No. DMA NAME Land Use Stormwater/ Mercury Person Population Region Pop Status DMA? Category 105 U.S. Army Federal Reservoir NA NA yes NWR/ Corps of WR Engineers 106 U.S. Bureau Federal Forestry NA NA no NWR/ of Land WR Management 107 U.S. Bureau Federal Reservoir NA NA yes NWR of Reclamation 108 U.S. Federal Other NA NA yes NWR/ Environment WR al Protection Agency 109 U.S. Fish & Federal Forestry NA NA no NWR/ Wildlife WR Service 110 U.S. Forest Federal Forestry NA NA no NWR/ Service WR 111 Clean Water Special Urban NA MS4 Phase 1 no NWR Services District 112 Eugene Special Reservoir NA NA no WR Water & District Electric Board 113 Metro Special Other NA NA no NWR District 114 Oak Lodge Special Urban NA MS4 Phase 1 yes NWR Water District Services District 115 Portland Special Reservoir NA NA ? NWR General District Electric 116 Port of Special Other NA MS4 Phase 1 no NWR Portland District 117 Tualatin Hills Special Other NA NA yes NWR Park and District Recreation District 118 Water Special Urban NA MS4 Phase 1 no NWR Environment District Services 119 Oregon Dept. State Agricultur NA NA no NWR/ of Agriculture e WR 120 Oregon Dept. State Other NA NA no NWR/ of Env. WR Quality 121 Oregon Dept. State Other NA NA yes NWR/ of Fish and WR Wildlife 122 Oregon Dept. State Forestry NA NA no NWR/ of Forestry WR

Oregon Department of Environmental Quality 161 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

DMA and Municipal New Responsible 2018 DEQ No. DMA NAME Land Use Stormwater/ Mercury Person Population Region Pop Status DMA? Category 123 Oregon Dept. State Other NA MS4 Phase 1 no NWR/ of WR Transportatio n 124 Oregon Dept. State Other NA NA no NWR/ State Lands WR 125 Oregon State Other NA NA yes NWR/ Dept.Geolog WR y & Mineral Ind. 126 Oregon State Other NA NA yes NWR/ Marine Board WR 127 Oregon State Other NA NA no NWR/ Parks and WR Recreation Department 128 Ash Creek Water Transport NA NA yes WR Water Conveyance Water Control District (Polk Co.) 129 Cedar Creek Water Transport NA NA yes WR Irrigation Conveyance Water District (Lane Co.) 130 Clackamas Water Transport NA NA yes NWR Bend Water Conveyance Water Control District (Clackamas Co) 131 Clackamas Water Transport NA NA yes NWR River Water Conveyance Water District (Clackamas Co) 132 Cloverdale Water Transport NA NA yes WR Water Conveyance Water Control District (Lane Co) 133 Columbia Water Transport NA NA yes NWR County Conveyance Water Drainage District #1 (Columbia Co) 134 Columbia Water Transport NA NA yes NWR Drainage Conveyance Water District No. 1 and No. 9

Oregon Department of Environmental Quality 162 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

DMA and Municipal New Responsible 2018 DEQ No. DMA NAME Land Use Stormwater/ Mercury Person Population Region Pop Status DMA? Category (Multnomah Co) 135 Cove Water Transport NA NA yes WR Orchard Conveyance Water Water Association (Yamhill Co) 136 Creswell Water Transport NA NA yes WR Irrigation Conveyance Water Association (Lane Co) 137 Creswell Water Transport NA NA yes WR Water Conveyance Water Control District (Lane Co) 138 Deer Island Water Transport NA NA yes NWR Drainage Conveyance Water Improvement (Clackamas Co) 139 Drainage Water Transport NA NA yes NWR District No. 8 Conveyance Water (Washington Co) 140 East Valley Water Transport NA NA no WR Water District Conveyance Water (Marion Co) 141 Fertile Water Transport NA NA yes WR Improvement Conveyance Water District (Lane Co) 142 G A Miller Water Transport NA NA yes WR Drainage Conveyance Water District No 1 (Marion Co) 143 Grand Prairie Water Transport NA NA yes WR Water Conveyance Water Control District (Linn Co) 144 Hawn Creek Water Transport NA NA yes WR District Conveyance Water Improvement Co. (Yamhill Co) 145 Job's Water Transport NA NA yes NWR Drainage Conveyance Water District (Washington Co)

Oregon Department of Environmental Quality 163 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

DMA and Municipal New Responsible 2018 DEQ No. DMA NAME Land Use Stormwater/ Mercury Person Population Region Pop Status DMA? Category 146 Junction City Water Transport NA NA yes WR Water Conveyance Water Control District (Lane Co) 147 Lacomb Water Transport NA NA yes WR Irrigation Conveyance Water District (Linn Co) 148 Lake Labish Water Transport NA NA no WR Water Conveyance Water Control District (Marion Co) 149 Lower Water Transport NA NA yes NWR Clackamas Conveyance Water River Water Control District 3 (Clackamas Co) 150 McKay Creek Water Transport NA NA yes NWR Water Conveyance Water Control District 2 (Washington Co) 151 Molalla River Water Transport NA NA yes NWR Water Conveyance Water Control District 3 (Clackamas Co) 152 Muddy Water Transport NA NA yes WR Creeks Conveyance Water Irrigation District (Linn Co) 153 Multnomah Water Transport NA NA yes NWR County Conveyance Water Drainage District 1 (Multnomah Co) 154 Murray, Water Transport NA NA yes NWR Smith & Conveyance Water Associates, Inc. (Multnomah Co) 155 North Water Transport NA NA yes WR Lebanon Conveyance Water Water

Oregon Department of Environmental Quality 164 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

DMA and Municipal New Responsible 2018 DEQ No. DMA NAME Land Use Stormwater/ Mercury Person Population Region Pop Status DMA? Category Control District (Linn Co) 156 Palmer Water Transport NA NA yes WR Creek Water Conveyance Water District Improvement Co. (Yamhill Co) 157 Peninsula Water Transport NA NA yes NWR Drainage Conveyance Water District No. 1 and No. 2 (Multnomah Co) 158 Queener Water Transport NA NA yes WR Irrigation Conveyance Water Improvement District (Linn Co) 159 Ranier Water Transport NA NA yes NWR Drainage Conveyance Water Improvement Company (Columbia Co.) 160 Rock Creek Water Transport NA NA yes WR Water District Conveyance Water (Polk/Yamhill Co) 161 Sandy Water Transport NA NA yes NWR Drainage Conveyance Water Improvement District No. 1 (Multnomah Co) 162 Santiam Water Transport NA NA yes WR Water Conveyance Water Control District (Marion Co) 163 Sauvie Island Water Transport NA NA yes NWR Drainage Conveyance Water District (Multnomah Co) 164 Scappoose Water Transport NA NA yes NWR Drainage Conveyance Water Improvement Company (Columbia Co)

Oregon Department of Environmental Quality 165 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

DMA and Municipal New Responsible 2018 DEQ No. DMA NAME Land Use Stormwater/ Mercury Person Population Region Pop Status DMA? Category 165 Scio Water Water Transport NA NA yes WR Improvement Conveyance Water District (Linn Co) 166 Shady Dell Water Transport NA NA yes NWR Water Conveyance Water Control District (Clackamas Co) 167 Sidney Water Transport NA NA yes WR Irrigation Conveyance Water District (Marion Co) 168 South Water Transport NA NA yes WR Santiam Conveyance Water River Water Control District (Marion Co) 169 Tualatin Water Transport NA NA yes NWR Valley Conveyance Water Irrigation District (Washington Co) 170 Washington Water Transport NA NA yes NWR County Conveyance Water Drainage District No. 7 (Washington Co) 171 West Labish Water Transport NA NA yes WR Water Conveyance Water Control District (Marion Co) 172 Woodburn Water Transport NA NA yes WR Hubbard Conveyance Water Drainage District (Marion Co)

Oregon Department of Environmental Quality 166 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 Appendix F. Oregon permitted mercury air emissions

Table F-1. 2016 Reported Mercury Emissions from Permitted Sources in the Willamette Basin Source 2016 Mercury Source Name Number Emissions [kg/yr] 36-5034 Cascade Steel Rolling Mills 22.889 24-5398 Covanta Marion, Inc. 2.984 22-3501 Cascade Pacific Pulp, LLC 1.297 26-3021 American Petroleum Environmental Services 0.576 22-6002 Freres Lumber Co. Inc. Lyons Facility 0.467 22-3010 Weyerhaeuser Company - Foster Engineered Lumber Products 0.317 34-2066 Stimson Lumber Company - Forest Grove Facility 0.281 02-2298 Oregon State University 0.222 22-2525 Frank Lumber Co., Inc. 0.222 26-1876 Owens-Brockway Glass Container Inc. 0.209 26-1865 EVRAZ Portland - Rivergate 0.177 34-2681 Intel Corporation 0.163 02-2173 Hollingsworth & Vose Fiber Company 0.136 22-2522 Freres Lumber Plant # 3 Plywood Division 0.136 22-0143 Flakeboard America Limited, Duraflake 0.127 26-2068 ESCO Corporation 0.082 22-1034 Bear Mountain Forest Products, Inc. 0.073 34-0157 T5 Datacenters 0.073 26-2050 Oregon Health and Sciences University 0.068 36-8031 Boise Cascade - Willamina Veneer 0.059 05-1849 Cascades Tissue Group - Oregon 0.054 22-0547 TDY Industries, LLC dba ATI Specialty Alloys and Components 0.054 22-6024 ENTEK International LLC 0.054 26-0088 Mutual Materials Company 0.054 26-1867 PCC Structurals, Inc., LPC 0.050 02-9502 Valley Landfills, Inc. 0.041 22-6034 Georgia-Pacific Consumer Products LP - Halsey Mill 0.041 36-0011 Riverbend Landfill 0.041 03-2533 Interfor US Inc. – Molalla Division 0.036 26-3009 Arclin Portland Division 0.031 03-1791 Sanders Wood Products/RSG 0.027 22-0328 Oregon Metallurgical, LLC dba ATI Albany Operations 0.027 03-0011 Clackamas County Tri-City Water Pollution Control Plant 0.023 26-2204 The Boeing Company 0.023 26-2968 Mondelez Global LLC 0.022

Oregon Department of Environmental Quality 167 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Source 2016 Mercury Source Name Number Emissions [kg/yr] 26-3240 Microchip Technology Inc. 0.022 26-3310 Metropolitan Service District. Saint Johns Landfill 0.022 24-5155 Oregon State Penitentiary 0.020 24-0060 Snyder's Lance Inc. 0.019 03-0010 Clackamas County Kellogg Creek Water Pollution Control Plant 0.018 03-0020 PCC Structurals, Inc., Deer Creek 0.018 03-2674 PCC Structurals, Inc., SSBO 0.018 26-0027 On Semiconductor 0.018 26-2914 Port of Portland 0.017 34-0067 CoorsTek Oregon Operations 0.017 26-3048 Oil Re-Refining Company 0.016 24-7067 NORPAC Foods, Inc. Stayton Plant #1 0.015 26-3067 Owens Corning Roofing and Asphalt, LLC 0.015 27-0012 Meduri Farms, Inc. 0.015 34-0010 SolarWorld Americas, Inc. 0.015 26-1815 Owens Corning Roofing & Asphalt, LLC 0.014 26-1894 Herbert Malarkey Roofing Company 0.014 34-9514 Agilyx Corporation 0.014 26-3267 U.S. Bancorp Columbia Center 0.013 34-0004 Hillsboro Landfill 0.013 26-2557 Blasen & Blasen Lumber Corp. 0.012 26-2952 United States Bakeries dba as Franz Bakery 0.011 26-0100 Columbia Boulevard WWTP 0.011 03-2624 Blount, Inc. 0.009 22-0011 Pacific Cast Technologies, Inc. dba ATI Cast Products 0.009 22-8045 OFD Foods, LLC - Plant 2 & 3 0.009 22-8056 EnerG2 Technologies, Inc. 0.009 26-1891 Ash Grove Cement Company - Rivergate 0.009 26-2197 Daimler Trucks North America, LLC. 0.009 26-3224 Vigor Industrial, LLC 0.009 34-0063 Lam Research 0.009 34-2804 Maxim Integrated 0.009 26-2390 Supreme Perlite Company 0.008 26-3291 The Boeing Company 0.008 34-2783 Bimbo Bakeries USA, Inc. 0.008 34-2813 Jireh Semiconductor Inc. 0.008 34-2638 Tektronix, Inc. 0.008 26-2777 Graphic Packaging International, Inc. 0.007 26-3135 Bullseye Glass Company 0.007 26-3241 Sapa Extrusions 0.007

Oregon Department of Environmental Quality 168 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Source 2016 Mercury Source Name Number Emissions [kg/yr] 36-7004 McFarland Cascade Holdings Inc. 0.007 26-2025 Arc Terminals Holdings, LLC 0.007 26-3051 International Paper- Portland Container 0.007 34-0055 Qorvo US 0.006 Boise Packaging & Newsprint, L.L.C. (subsidiary of Packaging 24-8061 0.006 Corporation of America) 26-2944 Gunderson, LLC. 0.006 22-8050 Stahlbush Island Farms, Inc. 0.005 26-2043 CertainTeed Corporation 0.005 34-0009 International Paper 0.005 26-1885 Galvanizers Company 0.005 26-2832 Portland State University 0.005 03-0098 PECO, Inc. 0.005 03-2634 Johnson Controls Battery Group, Inc. 0.005 03-2738 Consolidated Metco Inc. 0.005 03-2754 Safeway Clackamas Bread Plant 0.005 22-1024 Georgia-Pacific Chemicals LLC - Albany 0.005 26-2027 Chevron Products Company Willbridge Terminal 0.005 34-2678 Viasystems Corporation (dba TTM Technologies, Inc.) 0.005 26-2026 Phillips 66 Company 0.004 34-0005 Valmont Industries, Inc. 0.004 03-0050 Xerox 0.004 03-0099 Carver Readiness Center 0.004 26-3002 Siltronic Corporation 0.004 27-8034 Forest River, Inc. 0.003 27-0004 Forest River, Inc. 0.003 03-0093 Dave's Killer Bread 0.003 03-2505 Orchid Othropedic Solutions 0.003 34-2753 Rock Creek Advanced Wastewater Treatment Facility 0.003 34-9510 Summit Natural Energy 0.003 36-9504 City of Newberg WWTP 0.003 24-8062 Foster Farms, LLC - Donald Feedmill 0.003 34-2790 Tokyo Ohka Kogyo America, Inc. 0.003 22-6029 Panolam Industries Inc. 0.003 26-2572 Container Management Services, LLC 0.002 26-3272 Oldcastle APG West, Inc. dba Sakrete of the Pacific Northwest 0.002 02-9503 Pacific Northwest Generating Cooperative 0.002 22-6009 W. R. Grace 0.002 26-9550 Portland Service Center 0.002 24-0070 Specialty Polymers, Inc. 0.002 27-0005 Elkay Wood Products Company 0.002

Oregon Department of Environmental Quality 169 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Source 2016 Mercury Source Name Number Emissions [kg/yr] 03-2719 Georgia-Pacific Gypsum, LLC 0.001 22-8043 Forest River, Inc. 0.001 34-2623 Durham Facility 0.001 26-1869 Columbia Steel Casting Co., Inc. 0.001 BP West Coast Products, LLC: Portland Terminal - Seaport 26-2030 0.001 Midstream partners 22-8041 Selmet Inc. 0.001 24-9213 Panasonic Eco Solutions Solar America, LLC 0.001 36-8010 Hampton Lumber Mills, Inc. dba Willamina Lumber Company 0.001 03-0037 J and D Fertilizers, Ltd. 0.001 34-0058 International Paper Co. 0.001 26-1889 J.R. Simplot Company, Rivergate Terminal 0.001 34-0149 Avery Regional Service Center 0.001 26-2492 Northwest Pipe Company 0.001 26-3230 Lacamas Laboratories 0.001 24-0031 Superior Tire Service, Inc. 0.001 24-0136 City of Salem - Willow Lake WPCF 0.001 26-9554 Signature Graphics, Inc. 0.001

Oregon Department of Environmental Quality 170 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 Appendix G. Draft of Willamette River Instream Surrogate TSS-THg Analysis

By: Janani Govindan, Dan Sobota, and Kevin Brannan November 2019

Executive Summary

The Oregon Department of Environmental Quality (ODEQ) requires that Total Maximum Daily Load (TMDL) targets for mercury concentrations are meet in the Willamette River. An analysis was conducted using regression models to assess how Total Suspended Solid (TSS) concentrations could be used as a surrogate measure for instream total mercury (THg) concentrations in the Willamette River Basin. The dataset used for the modeling comprised of 63 surrogate TSS-THg samples with mercury concentrations above detection limits that were collected from 17 different sites that were dispersed throughout 9 HUC 8 subbasins within the Willamette River Basin. The following models were conducted for the analysis: An Ordinary Least Squares (OLS) model, a Linear mixed effects (LME) model with dry/wet seasons as a fixed effect and sites as a random effect, and an LME model with sites (excluding dry/wet seasons as a fixed effect). The results for the OLS model indicated that only 32% of the variance in THg concentrations could be explained by differences in TSS concentrations. The OLS model results suggested that other factors, such as spatial and temporal patterns, need to be taken into account for the analysis. The results for the LME model with dry/wet seasons as a fixed effect indicated that 80% of the variance in THg concentrations within the Willamette River could be explained by differences in TSS concentrations and by differences in site location, while 81% of the variance was explained by the LME model with sites as a random effect. Overall, the results suggest that spatial patterns such as site locations play a more significant role than temporal patterns such as seasonal differences (dry/wet periods) in the analysis. Thus, the LME model with sites was selected for the instream analysis.

Recommendations based on the analysis are listed below:

1) The LME model with sites should be used as a basis for establishing guidelines for the Willamette Basin Mercury TMDL 2) The LME equation can be used to find estimated TSS concentrations and percent reductions of THg concentrations

Oregon Department of Environmental Quality 171 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

1.1 Introduction The Oregon Department of Environmental Quality (ODEQ) has established a Total Maximum Daily Load (TMDL) for total mercury (THg) in the Willamette River and its tributaries to restore the beneficial use of fish consumption. The Willamette Basin Mercury TMDL was established to reduce concentrations of mercury to a level that no longer poses to be an unacceptable health risk to humans and aquatic wildlife. The Willamette Basin Mercury TMDL establishes acceptable loads (nonpoint sources) and wasteloads (point sources) of total mercury in the Willamette River Basin. The ODEQ has established an instream TMDL target THg concentration of 0.14 ng/L within the Willamette River Basin based on the simulated bioaccumulation of methylmercury for Northern Pikeminnow (ODEQ, 2019). However, because monitoring for total mercury can be difficult and cost-prohibitive, agencies and monitoring groups often use total suspended solids (TSS) as a surrogate for THg in stream. A strong relationship was found between TSS and THg in urban catchments (Eckley & Branfireun, 2009). In addition, TSS was used as a surrogate for DDT (Dichlorodiphenyltrichloroethane) to meet instream targets for the Lower Yakima TMDL in Washington (Johnson, 2007). A positive correlation between TSS and THg due to the capacity of THg to bind to particulate matter makes TSS a useful surrogate measure (Eckley & Branfireun, 2008). Therefore, provided that there is correlation between TSS and THg within the mainstem Willamette River and its tributaries, TSS could be used to predict instream THg concentrations in the Willamette River Basin (Hope & Rubin, 2004).

There are several major sources of THg concentrations within the Willamette River, including atmospheric deposition, soil/sediment erosion, and point source discharges from industrial facilities and stormwater outfalls. However, because controlling the sources of THg in atmospheric deposition is not tenable, DEQ has focused on controlling processes that increase of instream THg concentrations, including soil/sediment erosion (ODEQ, 2019).

The Environmental Protection Agency (EPA) and DEQ are using the watershed model Hydrological Simulation Program-FORTRAN (HSPF) to examine THg sources in surface water, subsurface systems, and soil/sediment erosion. Overall, this model can be used to assess the mass transport of THg from different land uses throughout the Willamette River Basin (TetraTech, 2019). For particulate THg, the HSPF model uses soil erosion rates simulated by the model paired with soil-THg data (weight of THg per weight of eroded soil) to calculate the mass transport of particulate THg to set particulate THg loads (TetraTech, 2019). Thus, the HSPF model provides calculations of the distribution of mercury in the water column and sediment. 1.2 Methods TetraTech, contracted by the Environmental Protection Agency (EPA), provided the paired surrogate data (TSS and THg concentrations) used in the analysis. Data had been compiled from the Willamette River Basin Mercury database (WRB Hg database) and from the Water Quality Portal (WQP) (TetraTech, 2018). There were 187 instream TSS-THg samples collected along the Willamette River main stem (from Portland to Cottage Grove) and tributaries leading into the Willamette River (Tualatin River, Clackamas River, North Santiam River, etc.) (Figure G-1 and Figure G-2). Samples of total mercury concentrations collected along the Portland Harbor Superfund site were high probably because of decades of industrial contamination along the site. Also, the influence of tidal exchange may cause excess mercury to exchange with sediments differently than in a unidirectional flow environment. To avoid combining paired data from different aquatic environments, all of the surrogate samples collected from the Portland Harbor Superfund site were omitted from the dataset, resulting in a total of 185 samples. Additionally, non-detect (censored) data were excluded from the analysis, resulting in a total of

Oregon Department of Environmental Quality 172 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

63 instream surrogate samples with detected mercury concentrations from 17 sampling locations (see discussion of methods for non-detect data in section 1.2.1).

All data were extracted from the WRB Hg database from 1/1/2002 to present. The 63 paired samples were collected from 9 different HUC 8 subbasins within the Willamette River Basin (Figure G-3). All of the 63 surrogate pairs had mercury concentrations with a detection limit of 0.5 ng/L (EPA Method 1631), which is above the instream target of 0.14 ng/L THg set for the Willamette. All of the samples used for the analysis were above this detection limit.

Each pair of surrogate TSS-THg data consisted of 1 TSS (mg/L) and 1 THg (ng/L) concentration that was collected at the same time and date. The sites (sampling locations) used in the analysis were defined by the latitude and longitude coordinates where the samples were collected.

The following methods were used:

• An initial exploratory analysis was used to provide summary statistics of TSS and THg and insights on transformation needed. • Box-Cox transformations were used to determine the most suitable power transformation to normalize the error distribution in the response variable. • Linear models were used to quantify the relationship between TSS and THg: – Ordinary Least Square (OLS) model – Linear Mixed Effects (LME) model with sites as a random effect and dry/wet seasons as a fixed effect – Linear Mixed Effects (LME) model with sites as a random effect The LME model was used to account for differences in site specific variation in parameters, and the addition of dry/wet seasons as a fixed effect was used to account for how dry/wet seasons impact the change in mean THg concentrations as seasons are one of the big drivers of precipitation within the Willamette River Basin (ODEQ, 2019). If the surrogate samples were collected from the months of June through October then it was classified as a dry season surrogate sample and if the surrogate samples were collected from the months of November through May then it was classified as a wet season surrogate sample.

To assess if the model sufficiently described the relationship between THg and TSS in the Willamette River, a diagnostic check of normality, constant variance, and independence was conducted for the two models. All analyses were conducted using R Version 3.5, Microsoft Office Excel (2016), and ArcMap version 10.5.1. 1.2.1 Addressing Non-Detect (Censored) data The dataset contained 65% of non-detect (censored) data, as there were 122 TSS-THg surrogate pairs with non-detected mercury concentrations within the original dataset of 187 surrogate pairs that TetraTech had provided DEQ. A total of five method detection limits (MDLs) were reported for the THg concentrations (0.5 ng/L, 20 ng/L, 30 ng/L, 40 ng/l, and 80 ng/l). There are several statistical techniques, such as the Maximum Likelihood Estimation (MLE) and the Regression on Order Statistics (ROS),that can handle non-detect data with multiple detection limits and can used to estimate the distribution of non-detects and determine summary statistics such as the mean and standard deviation of THg concentrations (Helsel, 2005). However, the larger the magnitude of the detection (censoring) limits, the more uncertainty and

Oregon Department of Environmental Quality 173 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

information loss when it comes to estimating model parameters (ITRC, 2013). Therefore, with a significant difference in the order of magnitude between the detection limits (i.e. 0.5 ng/L versus 20, 30, 40, or 80 ng/L), the decision was made to remove the 122 non-detects from the dataset. A total of 63 surrogate pairs with detected THg concentrations were used for the instream analysis and all of the pairs had THg concentrations above their respective detection limit of 0.5 ng/L. 1.2.2 Ordinary Least Squares The ordinary least squares (OLS) model was constructed using TSS as the independent variable (fixed effect) and THg as the dependent variable. The OLS model assumes normality in parameter estimates and in the distribution of residuals. THg Concentrations, as a function of TSS concentrations, were generated using the following equation:

{Y|X}= + X where,

Y (Response𝛽𝛽0 𝛽𝛽 Variable)1 = log10 transformed THg concentrations (ng/L) X (Predictor Variable) = log10 transformed TSS concentrations (mg/L) = Intercept

𝛽𝛽0 = Slope (Rate of change in log10 transformed THg concentrations as a function of the change in log10 transformed TSS concentrations) 𝛽𝛽1 1.2.3 Linear Mixed-Effects Model The LME model is similar to the OLS model, as it also uses the dependent variable (THg concentrations) and the fixed effect (TSS concentrations), but it also includes a random effects variable (sites). Random effects and fixed effects are both explanatory variables. However, random effects explain the change in the variance of the dependent variable and the fixed effect explains the change in the mean of the dependent value. In other words, we expect that mean THg concentrations will increase with TSS concentrations across all sites, but that the variance around the increase may depend on the given site.

Two LME models were used for the analysis, with both incorporating sites as a random effect. One LME model included dry/wet seasons as an additional fixed effect while the second LME model excluded the fixed effect of dry/wet seasons. By including dry/wet seasons as a fixed effect, we expect that the mean THg concentrations will increase or decrease according to whether the samples were collected during dry (months of June-October) or wet (months of November-May) seasons. Thus, in order to test the hypothesis that THg will increase with TSS and that mean THg concentrations will increase or decrease according to whether it is a dry or wet season, both site and seasonal variation should be taken into account using the mixed effects model approach.

A random intercept LME model was fitted to the data to model differences between sites within the same data set. A total of 63 paired samples were split into 17 groups of sites. The LME model used the following general equation (Helwig, 2017):

µ{Y|X1, X2, V} = + + + ( + ~N(0, )) where, 2 Y (Response Variable)𝛽𝛽0 𝛽𝛽1𝑋𝑋 =1 log10𝛽𝛽2𝑋𝑋2 transformed𝛴𝛴 𝑉𝑉𝑖𝑖 𝜖𝜖 THg𝜎𝜎 concentrations (mg/L) (Predictor Variable/Fixed Effect) = log10 transformed TSS concentrations (mg/L)

𝑋𝑋1 (Additional Fixed Effect) = seasons (dry/wet) 𝑋𝑋2 Oregon Department of Environmental Quality 174 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

(intercepts of each random effect) = sampling sites

𝑉𝑉𝑖𝑖 = Intercept 𝛽𝛽0 = Slope (Rate of change in log10 transformed THg concentrations as a function of the change in log10 transformed TSS concentrations) 𝛽𝛽1 = Slope (Rate of change in log10 transformed THg concentrations as a function of change in seasons (dry/wet) 𝛽𝛽2 = Error of the residuals (where N represents residuals that are normally distributed with a mean2 of zero and have a variance of ) 𝜎𝜎 2 1.2.4 Prediction Intervals𝜎𝜎 A 95 percent prediction interval was used to calculate a likely range of future THg values based on the LME models. 1.3 Results The plots demonstrate a right-skewed distribution of sampling data (Figure G-4). The skewed data were transformed to improve normality of the data distribution, which is needed to meet that assumptions for the OLS model. The variation in TSS-THg concentrations across all HUC 8 subbasins and within each individual HUC 8 subbasin was examined (Table G-3, Table G-4, and Table G-5). Scatterplots were used to illustrate the variation in TSS-THg Concentrations across all HUC 8 subbasins and across dry/wet seasons (Figure G-5, Figure G-6, and Figure G-7).The scatterplots indicate that there is a stronger correlation between TSS and THg concentrations within the Upper Willamette, Middle Willamette, and Clackamas subbasins and that THg and TSS concentrations tend to be higher in the wet season. As determined by the box-cox transformation, a log10 transformation was applied to the independent (predictor) and dependent (response) variables for development of the OLS regression (Figure G-8).

The OLS model showed a significant (p < 6.62 x 10-8) correlation between THg and TSS concentrations. However, the OLS model provides a weak fit to the data as there is a large difference between the values of THg sample concentrations and the THg concentrations predicted by the OLS regression (Figure G-9). An adjusted R-squared value of 0.372 indicates that TSS concentrations alone does explain a large amount of variation in THg concentrations (Table G-6). The following OLS equation was determined:

(1) (THg conc.) = 0.412 x (TSS conc.) - 0.062

10 10 The LME𝑙𝑙𝑙𝑙𝑔𝑔 model, with sites as a random𝑙𝑙𝑙𝑙𝑔𝑔 effect and the addition of dry/wet seasons as a fixed effect showed a significant (p < 3.68 x 10-10) positive correlation between THg and TSS concentrations (Table G-7, Figure G-10). The conditional (fixed and random effects added together) R-squared value was 0.80 versus 0.465 for the marginal (fixed effects only) R-squared value (). The variance of THg due to TSS increased greatly with the addition of the random effects variable (Section G-1.2.3). The following equation which resulted from the LME model was used:

(2) (THg conc.) = 0.449 x (TSS conc.) + 0.064 x (Seasons) - 0.086

10 10 The LME𝑙𝑙𝑙𝑙𝑔𝑔 model without the fixed effect𝑙𝑙𝑙𝑙𝑔𝑔 of dry/wet seasons performed slightly better than the LME model with seasons (Table G-9, Figure G-11). The conditional R-squared value of the model (0.81) was slightly

Oregon Department of Environmental Quality 175 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

bigger than the conditional R-squared value from the LME model including seasons (0.80) (). The variance of THg due to TSS increased greatly with the addition of the random effect in both LME models.

(3) (THg conc.) = 0.506 x (TSS conc.) - 0.089

𝑙𝑙𝑙𝑙 𝑔𝑔10 𝑙𝑙𝑙𝑙𝑔𝑔10 Overall, both LME models performed better in quantifying the correlation between THg and TSS concentrations than the OLS model. The 95 percent prediction intervals for the range of samples and model parameters are displayed in Figure G-10 and Figure G-11. The diagnostic statistics for the OLS model and LME model are presented in Figure G-12-14. Estimated TSS concentrations based on the LME model with sites as a random effect are listed in Table G-11, and percent reductions of THg are listed in Table G-12. In addition, a cumulative distribution function (cdf) plot of TSS sample concentrations for uncensored data is displayed in Figure G-15. 1.4 Discussion The OLS model results indicate that TSS concentrations could explain < 50% of variation in THg concentrations. Without taking sites or dry/wet seasons into account when developing the model, model error estimates may be too high for predictions. Additionally, the high uncertainty of the OLS model may result from spatial or temporal patterns in the data that needed to be taken into account within the model.

In order to account for spatial variability, a random effect of sites was incorporated into the LME models. To account for temporal variability, dry/wet seasons were included as a fixed effect in the LME model with sites as a random effect. Two LME models were fit to the sample data in order to quantify how much of the residual variation (variation in the error) in THg concentrations could be taken into account by the random effect of sites and to quantify if dry/wet seasons affected the mean of THg concentrations. Although both of the LME models performed well, the LME model which excluded dry/wet seasons as a fixed effect and included sites as a random effect performed slightly better than the LME model including dry/wet seasons as a fixed effect in explaining the variance in THg concentrations. Thus, the results of the analysis indicate that dry/wet seasons do not play a significant role in explaining the change in mean THg concentration values. These results also indicate that sites (sampling locations) play an important role in affecting variation in THg concentrations. Thus, the LME model with sites (excluding dry/wet seasons) was selected for the instream analysis. 1.5 Recommendations The LME model with sites (sampling locations) only can be used to calculate TSS concentrations that correspond to designated THg concentrations. For instance, the LME equation can be used to find estimated TSS concentrations and calculate percent reductions of THg concentrations (example applications are listed below). Based on the strong relationship found between total suspended solids and total mercury interim goals were set for reductions in TSS concentration to measure progress in the reduction in total mercury loads. The approach is to set reductions in the 75-percentile for the TSS concentration over time. The initial 75 percentile of TSS concentration is based on the 2019 dataset used to develop the TSS and total mercury concentrations. The schedule is listed in Table G-13. The 75th percentile for the total suspended solids concentrations can be seen in Figure G-16, which is the empirical cumulative distribution function for the observed total suspended solids concentrations. The empirical cumulative distribution function will be reanalyzed when new data becomes available and a

Oregon Department of Environmental Quality 176 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Bayesian inference approach to structured decision making (Conroy & Peterson, 2013) will be used to update the statistical distribution used to assess the progress in meeting the reduction targets and to reassess past targets if any had occurred.

LME Model Equation excluding seasons:

(THg conc.) (ng/L) = 0.506 x (TSS conc.) (mg/L) - 0.089

𝑙𝑙𝑙𝑙Example𝑔𝑔10 Application of LME Model𝑙𝑙𝑙𝑙𝑔𝑔10 excluding seasons:

(1) To find estimated TSS Concentrations (The range of THg concentrations were based on the THg summary statistics)

• Substitute 0.14 ng/L for the THg Concentration (instream THg target)

Log10 (0.14) = 0.506 x log10 (X) - 0.089 -0.853872= 0.506 x log10 (X) – 0.089 (-0.853872+ 0.089) = 0.506 x log10 (X) (-0.853872+ 0.089)/ 0.506 = log10 (X) -1.511605 = log10 (X) X = 0.031 TSS mg/L (2) To find percent reductions in THg concentrations:

Percent Reduction from 100 mg/L to 80 mg/L

• Substitute 100 mg/L for the TSS Concentration log10 (THg conc.) = 0.506 x log10 (100 mg/L) -0.089 log10 (THg conc.) = 0.923 10(0.923) = 8.38 ng/L

• Substitute 80 mg/L for the TSS Concentration log10 (THg conc.) = 0.506 x log10 (80 mg/L) -0.089 log10 (THg conc.) = 0.874 10(0.874) = 7.48 ng/L

• Use the following equation to find the percent reduction % Reduction = (Original value – New value)/ (Original value) x 100

(8.38-7.48)/ (8.38) x 100 = 10.7

Oregon Department of Environmental Quality 177 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

1.6 Tables Table G-1. Full TSS-THg Dataset Provided by TetraTech for the Instream Willamette River Basin Surrogate Analysis. THg Detection Excluded THg TSS Non- Reasons Latitude Longitude Date Limit from (ng/L) (mg/L) Detect Excluded (ng/L) Analysis (ND) PDX Harbor 45.5800 -122.7500 9/6/2002 187 4.4 NO NA YES Superfund Site 45.5800 -122.7500 11/8/2004 40 5 YES 40 YES ND THg 45.5700 -122.7200 11/10/2004 40 5 YES 40 YES ND THg 45.6200 -122.8000 11/12/2004 40 7 YES 40 YES ND THg 45.6300 -122.7800 11/12/2004 40 5 YES 40 YES ND THg 45.6300 -122.7900 11/12/2004 40 5 YES 40 YES ND THg 45.5400 -122.6800 11/15/2004 40 8 YES 40 YES ND THg 45.5800 -122.7600 11/17/2004 40 5 YES 40 YES ND THg 45.6100 -122.7800 11/18/2004 40 5 YES 40 YES ND THg 45.6000 -122.7700 11/19/2004 40 5 YES 40 YES ND THg 45.6100 -122.7900 11/19/2004 40 8 YES 40 YES ND THg 45.6100 -122.7800 11/22/2004 40 7 YES 40 YES ND THg 45.5900 -122.7600 11/23/2004 40 7 YES 40 YES ND THg 45.6000 -122.7700 11/23/2004 40 6 YES 40 YES ND THg 45.5800 -122.7500 11/29/2004 40 5 YES 40 YES ND THg 45.5700 -122.7400 11/30/2004 40 5 YES 40 YES ND THg 45.5600 -122.7100 12/1/2004 40 5 YES 40 YES ND THg 45.5700 -122.7100 12/1/2004 40 5 YES 40 YES ND THg 45.5700 -122.7400 12/1/2004 40 7 YES 40 YES ND THg 45.5800 -122.7400 12/1/2004 40 5 YES 40 YES ND THg 45.5900 -122.7700 12/1/2004 40 5 YES 40 YES ND THg 45.5500 -122.7100 12/2/2004 40 5 YES 40 YES ND THg 45.5600 -122.7200 12/2/2004 40 5 YES 40 YES ND THg 45.5800 -122.7600 12/2/2004 40 5 YES 40 YES ND THg 45.5800 -122.7500 3/1/2005 40 5 YES 40 YES ND THg 45.5700 -122.7200 3/3/2005 40 5 YES 40 YES ND THg 45.6100 -122.7900 3/4/2005 40 10 YES 40 YES ND THg 45.6200 -122.8000 3/4/2005 40 5 YES 40 YES ND THg 45.6300 -122.7800 3/4/2005 40 5 YES 40 YES ND THg 45.6300 -122.7900 3/4/2005 40 5 YES 40 YES ND THg 45.5400 -122.6800 3/7/2005 40 6.5 YES 40 YES ND THg 45.5900 -122.7600 3/9/2005 40 6 YES 40 YES ND THg 45.6100 -122.7800 3/10/2005 40 6 YES 40 YES ND THg 45.5900 -122.7700 3/11/2005 40 5 YES 40 YES ND THg 45.6000 -122.7700 3/11/2005 40 5.25 YES 40 YES ND THg 45.5800 -122.7500 3/14/2005 40 7 YES 40 YES ND THg

Oregon Department of Environmental Quality 178 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

THg Detection Excluded THg TSS Non- Reasons Latitude Longitude Date Limit from (ng/L) (mg/L) Detect Excluded (ng/L) Analysis (ND) 45.5700 -122.7400 3/15/2005 40 5 YES 40 YES ND THg 45.5500 -122.7100 3/16/2005 40 9 YES 40 YES ND THg 45.5600 -122.7100 3/16/2005 40 5 YES 40 YES ND THg 45.5600 -122.7200 3/16/2005 40 14 YES 40 YES ND THg 45.5700 -122.7100 3/16/2005 40 5 YES 40 YES ND THg 45.5700 -122.7400 3/16/2005 40 10 YES 40 YES ND THg 45.5800 -122.7400 3/17/2005 40 6 YES 40 YES ND THg 45.5800 -122.7600 3/17/2005 40 5 YES 40 YES ND THg 45.6100 -122.7800 3/17/2005 40 7 YES 40 YES ND THg 45.3400 -122.6500 5/23/2005 2.17 17 NO 0.5 NO NA 45.6100 -122.7800 7/5/2005 80 9 YES 80 YES ND THg 45.6200 -122.8000 7/5/2005 80 15 YES 80 YES ND THg 45.6300 -122.7800 7/5/2005 80 7 YES 80 YES ND THg 45.6300 -122.7900 7/5/2005 80 13 YES 80 YES ND THg 45.5800 -122.7500 7/6/2005 80 8 YES 80 YES ND THg 45.5900 -122.7700 7/8/2005 80 10 YES 80 YES ND THg 45.6000 -122.7700 7/8/2005 80 5.5 YES 80 YES ND THg 45.6100 -122.7900 7/8/2005 80 7 YES 80 YES ND THg 45.5400 -122.6800 7/11/2005 80 17 YES 80 YES ND THg 45.5800 -122.7600 7/12/2005 80 25 YES 80 YES ND THg 45.6100 -122.7800 7/13/2005 80 8 YES 80 YES ND THg 45.5700 -122.7200 7/14/2005 80 4 YES 80 YES ND THg 45.5700 -122.7400 7/15/2005 80 10 YES 80 YES ND THg 45.5800 -122.7400 7/15/2005 80 5 YES 80 YES ND THg 45.5800 -122.7600 7/15/2005 80 13 YES 80 YES ND THg 45.5900 -122.7600 7/15/2005 80 7 YES 80 YES ND THg 45.5700 -122.7400 7/18/2005 80 7 YES 80 YES ND THg 45.5800 -122.7500 7/19/2005 80 5 YES 80 YES ND THg 45.5500 -122.7100 7/20/2005 80 8 YES 80 YES ND THg 45.5600 -122.7100 7/20/2005 80 3 YES 80 YES ND THg 45.5600 -122.7200 7/20/2005 80 12 YES 80 YES ND THg 45.5700 -122.7100 7/20/2005 80 4 YES 80 YES ND THg 45.4700 -122.6700 1/19/2006 80 49 YES 80 YES ND THg 45.5400 -122.6800 1/20/2006 80 62 YES 80 YES ND THg 45.6100 -122.7800 1/21/2006 80 49 YES 80 YES ND THg 45.6300 -122.7900 9/4/2006 20 7 YES 20 YES ND THg 45.5400 -122.6800 9/5/2006 20 1 YES 20 YES ND THg 45.6300 -122.7900 9/5/2006 20 4 YES 20 YES ND THg 45.5400 -122.6800 9/6/2006 20 5 YES 20 YES ND THg 45.5400 -122.6900 9/6/2006 20 10 YES 20 YES ND THg 45.6200 -122.8000 9/7/2006 20 9 YES 20 YES ND THg

Oregon Department of Environmental Quality 179 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

THg Detection Excluded THg TSS Non- Reasons Latitude Longitude Date Limit from (ng/L) (mg/L) Detect Excluded (ng/L) Analysis (ND) 45.6100 -122.7800 9/8/2006 20 12 YES 20 YES ND THg 45.5800 -122.7600 9/12/2006 20 10.5 YES 20 YES ND THg 45.4700 -122.6700 9/13/2006 20 3 YES 20 YES ND THg 45.5400 -122.6800 11/2/2006 25 2.5 YES 20 YES ND THg PDX Harbor 45.5400 -122.6900 11/2/2006 30 2 NO 20 YES Superfund Site 45.6300 -122.7900 11/2/2006 20 4.5 YES 20 YES ND THg 45.4700 -122.6700 11/3/2006 20 4.25 YES 20 YES ND THg 45.5700 -122.7500 11/3/2006 20 5.25 YES 20 YES ND THg 45.5800 -122.7600 11/3/2006 20 3.5 YES 20 YES ND THg 45.6000 -122.7800 11/3/2006 20 4.5 YES 20 YES ND THg 45.6300 -122.7900 11/3/2006 20 5 YES 20 YES ND THg 45.5600 -122.7200 11/4/2006 20 4.5 YES 20 YES ND THg 45.5700 -122.7100 11/4/2006 20 4 YES 20 YES ND THg 45.5700 -122.7400 11/4/2006 20 3.5 YES 20 YES ND THg 45.5900 -122.7700 11/4/2006 20 3 YES 20 YES ND THg 45.6100 -122.7800 11/4/2006 20 5 YES 20 YES ND THg 45.6200 -122.8000 11/4/2006 20 3.5 YES 20 YES ND THg 45.6300 -122.7900 11/4/2006 20 5.5 YES 20 YES ND THg 45.5500 -122.7000 11/5/2006 20 4.5 YES 20 YES ND THg 45.5500 -122.7100 11/5/2006 20 3 YES 20 YES ND THg 45.5800 -122.7500 11/5/2006 20 3.5 YES 20 YES ND THg 45.5800 -122.7600 11/5/2006 30 3 YES 20 YES ND THg 45.4700 -122.6700 1/15/2007 30 12.5 YES 20 YES ND THg 45.5400 -122.6800 1/15/2007 40 13 YES 20 YES ND THg 45.6300 -122.7900 1/15/2007 40 12 YES 20 YES ND THg 45.5800 -122.7500 2/21/2007 20 15.75 YES 20 YES ND THg 45.5500 -122.7000 2/22/2007 20 15.6 YES 20 YES ND THg 45.5500 -122.7100 2/23/2007 20 21 YES 20 YES ND THg 45.5700 -122.7400 2/24/2007 20 17.5 YES 20 YES ND THg 45.5800 -122.7600 2/24/2007 20 36 YES 20 YES ND THg 45.5700 -122.7100 2/25/2007 20 15.25 YES 20 YES ND THg 45.5700 -122.7500 2/26/2007 20 21.5 YES 20 YES ND THg 45.6300 -122.7900 2/26/2007 20 14.25 YES 20 YES ND THg 45.5600 -122.7200 2/27/2007 20 27 YES 20 YES ND THg 45.6200 -122.8000 2/27/2007 20 33 YES 20 YES ND THg 45.5800 -122.7600 3/1/2007 20 22 YES 20 YES ND THg 45.5900 -122.7700 3/1/2007 20 18.25 YES 20 YES ND THg 45.6100 -122.7800 3/1/2007 20 19.5 YES 20 YES ND THg 45.5400 -122.6800 3/2/2007 20 23 YES 20 YES ND THg

Oregon Department of Environmental Quality 180 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

THg Detection Excluded THg TSS Non- Reasons Latitude Longitude Date Limit from (ng/L) (mg/L) Detect Excluded (ng/L) Analysis (ND) 45.5400 -122.6800 3/3/2007 20 18.5 YES 20 YES ND THg 45.6100 -122.7800 3/3/2007 20 16.75 YES 20 YES ND THg 45.5400 -122.6900 3/4/2007 20 21.5 YES 20 YES ND THg 45.6000 -122.7800 3/5/2007 20 17.5 YES 20 YES ND THg 45.6300 -122.7900 3/8/2007 20 10.5 YES 20 YES ND THg 45.6300 -122.7900 3/9/2007 20 10.5 YES 20 YES ND THg 45.6300 -122.7900 3/10/2007 20 10 YES 20 YES ND THg 44.2700 -123.1700 4/5/2010 2.95 9 NO 0.5 NO NA 44.5500 -123.2500 4/5/2010 2.95 11 NO 0.5 NO NA 44.7200 -123.0100 4/5/2010 1.06 4 NO 0.5 NO NA 44.9400 -123.0500 4/5/2010 3.06 20 NO 0.5 NO NA 45.0900 -123.0400 4/5/2010 2.77 11 NO 0.5 NO NA 43.6000 -122.4600 4/6/2010 0.968 1 NO 0.5 NO NA 43.9800 -122.9700 4/7/2010 5.26 5 NO 0.5 NO NA 44.0700 -123.1100 4/7/2010 3.61 5 NO 0.5 NO NA 44.1100 -123.0500 4/7/2010 1.11 4 NO 0.5 NO NA 44.0000 -122.9100 4/7/2010 1.36 2 NO 0.5 NO NA 45.4000 -122.5600 4/8/2010 1.88 15 NO 0.5 NO NA 45.3500 -122.6800 4/19/2010 3.61 8 NO 0.5 NO NA 45.2900 -122.9700 4/20/2010 1.44 6 NO 0.5 NO NA 45.3400 -122.6400 4/20/2010 1.56 5 NO 0.5 NO NA 45.3700 -122.6000 4/21/2010 1.42 6 NO 0.5 NO NA 45.5800 -122.7600 4/21/2010 1.39 4 NO 0.5 NO NA 44.1100 -123.0500 7/19/2010 0.5 2 YES 0.5 YES ND THg 44.2700 -123.1700 7/19/2010 1 2 NO 0.5 NO NA 44.5500 -123.2500 7/19/2010 1.12 3 NO 0.5 NO NA 44.7200 -123.0100 7/19/2010 0.651 3 NO 0.5 NO NA 43.6000 -122.4600 7/20/2010 0.5 1 YES 0.5 YES ND THg 43.9800 -122.9700 7/20/2010 2.1 2 NO 0.5 NO NA 44.0000 -122.9100 7/20/2010 1.1 1 NO 0.5 NO NA 44.0700 -123.1100 7/21/2010 1.24 2 NO 0.5 NO NA 45.3500 -122.6800 7/21/2010 1.4 6 NO 0.5 NO NA 45.4000 -122.5600 7/22/2010 0.811 2 NO 0.5 NO NA 44.9400 -123.0500 7/26/2010 0.819 3 NO 0.5 NO NA 45.0900 -123.0400 7/26/2010 0.792 2 NO 0.5 NO NA 45.2900 -122.9700 7/27/2010 0.82 5 NO 0.5 NO NA 45.3400 -122.6400 7/27/2010 0.726 1 NO 0.5 NO NA 45.3700 -122.6000 7/28/2010 0.944 4 NO 0.5 NO NA 45.5800 -122.7600 7/28/2010 0.896 6 NO 0.5 NO NA 44.9400 -123.0500 10/18/2010 0.892 4 NO 0.5 NO NA 45.0900 -123.0400 10/18/2010 0.999 5 NO 0.5 NO NA

Oregon Department of Environmental Quality 181 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

THg Detection Excluded THg TSS Non- Reasons Latitude Longitude Date Limit from (ng/L) (mg/L) Detect Excluded (ng/L) Analysis (ND) 45.2900 -122.9700 10/19/2010 0.869 3 NO 0.5 NO NA 45.3400 -122.6400 10/19/2010 0.886 2 NO 0.5 NO NA 45.5800 -122.7600 10/20/2010 1.24 4 NO 0.5 NO NA 45.3700 -122.6000 10/21/2010 0.826 2 NO 0.5 NO NA 43.6000 -122.4600 10/26/2010 2.27 1 NO 0.5 NO NA 43.9800 -122.9700 10/26/2010 5.17 10 NO 0.5 NO NA 44.0000 -122.9100 10/26/2010 1.72 3 NO 0.5 NO NA 44.0700 -123.1100 10/27/2010 3.1 5 NO 0.5 NO NA 44.1100 -123.0500 10/27/2010 1.65 5 NO 0.5 NO NA 44.2700 -123.1700 10/27/2010 2.37 6 NO 0.5 NO NA 44.5500 -123.2500 10/27/2010 2.35 8 NO 0.5 NO NA 44.7200 -123.0100 10/27/2010 0.959 3 NO 0.5 NO NA 45.3500 -122.6800 10/28/2010 2.82 8 NO 0.5 NO NA 45.4000 -122.5600 10/28/2010 1.84 2 NO 0.5 NO NA 45.3500 -122.6800 1/3/2011 4.01 15 NO 0.5 NO NA 45.4000 -122.5600 1/3/2011 1.41 3 NO 0.5 NO NA 44.0700 -123.1100 1/4/2011 3.34 7 NO 0.5 NO NA 44.1100 -123.0500 1/4/2011 0.847 3 NO 0.5 NO NA 44.2700 -123.1700 1/4/2011 2.41 10 NO 0.5 NO NA 43.6000 -122.4600 1/5/2011 0.622 1 NO 0.5 NO NA 43.9800 -122.9700 1/5/2011 4.04 3 NO 0.5 NO NA 44.0000 -122.9100 1/5/2011 1.85 2 NO 0.5 NO NA 44.5500 -123.2500 1/6/2011 2.33 6 NO 0.5 NO NA 44.7200 -123.0100 1/6/2011 0.934 4 NO 0.5 NO NA 44.9400 -123.0500 1/10/2011 1.96 9 NO 0.5 NO NA 45.0900 -123.0400 1/10/2011 1.89 8 NO 0.5 NO NA 45.5800 -122.7600 1/11/2011 1.79 6 NO 0.5 NO NA 45.2900 -122.9700 1/13/2011 3.18 65 NO 0.5 NO NA 45.3400 -122.6400 1/13/2011 2.77 15 NO 0.5 NO NA 45.3700 -122.6000 1/14/2011 3.45 30 NO 0.5 NO NA

Oregon Department of Environmental Quality 182 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Table G-2. Sampling Sites for the Instream Willamette River TSS-THg Surrogate Analysis Sampling Number of HUC 8 Site Samples Code HUC 8 Description Longitude Latitude 1 3 17090001 Middle Fork Willamette -122.46 43.60 2 4 17090001 Middle Fork Willamette -122.91 44.00 3 4 17090002 Coast Fork Willamette -122.97 43.98 4 3 17090003 Upper Willamette -123.11 44.07 5 4 17090003 Upper Willamette -123.17 44.27 6 4 17090003 Upper Willamette -123.25 44.55 7 4 17090004 Mckenzie -123.05 44.11 8 4 17090005 North Santiam -123.01 44.72 9 4 17090007 Middle Willamette -123.05 44.94 10 4 17090007 Middle Willamette -123.04 45.09 11 4 17090007 Middle Willamette -122.97 45.29 12 4 17090007 Middle Willamette -122.64 45.34 13 1 17090007 Middle Willamette -122.65 45.34 14 4 17090010 Tualatin -122.68 45.35 15 4 17090011 Clackamas -122.60 45.37 16 4 17090011 Clackamas -122.56 45.40 17 4 17090012 Lower Willamette -122.76 45.58

Table G-3. Summary Statistics for Untransformed TSS (mg/L) and THg Concentrations (ng/L) across all HUC 8 Subbasins TSS Concentrations (mg/L) THg Concentrations (ng/L) Min. 1.00 0.622 1st Qu. 3.00 0.964 Median 5.00 1.560 Mean 6.87 1.901 3rd Qu. 8.00 2.590 Max. 65.00 5.260

Table G-4. Summary Statistics for Untransformed THg Concentrations (mg/L) for each individual HUC 8 Subbasin HUC 8 Code Min. 1st Qu. Median Mean 3rd Qu. Max. 17090001 0.622 1.034 1.360 1.413 1.785 2.27 17090002 2.100 3.555 4.605 4.143 5.192 5.26 17090003 1.000 2.058 2.390 2.397 2.987 3.61 17090004 0.847 0.979 1.110 1.202 1.380 1.65 17090005 0.651 0.863 0.947 0.901 0.984 1.06 17090007 0.726 0.869 1.440 1.624 2.170 3.18 17090010 1.400 2.465 3.215 2.960 3.710 4.01 17090011 0.811 0.914 1.415 1.573 1.850 3.45 17090012 0.896 1.154 1.315 1.329 1.490 1.79

Oregon Department of Environmental Quality 183 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Table G-5. Summary Statistics for Untransformed TSS Concentrations (ng/L) for each individual HUC 8 Subbasin HUC 8 Code Min. 1st Qu. Median Mean 3rd Qu. Max. 17090001 1 1.00 1.0 1.57 2.00 3 17090002 2 2.75 4.0 5.00 6.25 10 17090003 2 4.50 6.0 6.17 8.25 11 17090004 3 3.50 4.0 4.00 4.50 5 17090005 3 3.00 3.5 3.50 4.00 4 17090007 1 3.00 5.0 10.65 11.00 65 17090010 6 7.50 8.0 9.25 9.75 15 17090011 2 2.00 3.5 8.00 8.25 30 17090012 4 4.00 5.0 5.00 6.00 6

Table G-6. OLS Model Summary Results for all HUC 8 Subbasins. 2 Adjusted R = 0.37; F1,61 = 37.78; p < 0.0001.

Estimate Std. Error t value Pr(>|t|) (Intercept) -0.0622 0.0506 -1.23 2.24e-01 TSS 0.4118 0.0670 6.15 1.00e-07

Table G-7. LME Model Summary Results for all HUC 8 Subbasins with Sites as a Random Effect and Dry/Wet Seasons as a Fixed Effect. Dry is the baseline model. Marginal R2 (value associated with fixed effects) = 0.47; conditional R2 (fixed plus random effects) = 0.80.

Effect Group/Term Estimate Std.Error Statistic df p.value Fixed (Intercept) -0.124 0.052 -2.4 34 2.3e-02 Fixed TSS 0.449 0.059 7.6 55 3.7e-10 Fixed Season (wet) 0.064 0.035 1.8 48 7.8e-02 Effect Group/Term Estimate Std.Deviation Random Site ID (Intercept) Variance 0.0238 0.154 Random Residual Variance 0.0138 0.118

Table G-8. Values for adjusting the intercept in the LME model for total mercury and TSS with season and site as random effects. Latitude Longitude Effect (intercept) 43.60 -122.46 0.106557196 43.98 -122.97 0.353995813 44.07 -123.11 0.195325545 44.11 -123.05 -0.100804005 44.27 -123.17 0.051508218 44.55 -123.25 0.041959716 44.72 -123.01 -0.176085328 44.94 -123.05 -0.105623812 44.00 -122.91 0.123063062 45.09 -123.04 -0.072762542 45.29 -122.97 -0.175926782 45.34 -122.64 -0.035334002

Oregon Department of Environmental Quality 184 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Latitude Longitude Effect (intercept) 45.34 -122.65 -0.097859162 45.35 -122.68 0.096080952 45.37 -122.60 -0.101380753 45.40 -122.56 -0.009645414 45.58 -122.76 -0.093068702

Table G-9. LME Model Summary Results for all HUC 8 Subbasins with Sites as a Random Effect but without the Fixed Effect of Dry/Wet Seasons. Marginal R2 (value associated with fixed effects) = 0.47; conditional R2 (fixed plus random effects) = 0.81. Effect Term Estimate Std.Error Statistic df p.value Fixed (Intercept) -0.128 0.054 -2.4 34 2.2e-02 Fixed log_fin_TSS 0.506 0.051 10.0 53 8.6e-14 Effect Estimate Estimate Std.Deviation Random Site ID (Intercept) Variance 0.0251 0.158 Random Residual Variance 0.0143 0.119

Table G-10. Values for adjusting the intercept in the LME model for total mercury and TSS with site as a random effect. Latitude Longitude Effect (intercept) 43.60 -122.46 0.145800940 43.98 -122.97 0.355304338 44.07 -123.11 0.195298960 44.11 -123.05 -0.090824059 44.27 -123.17 0.044918277 44.55 -123.25 0.033262624 44.72 -123.01 -0.172463553 44.94 -123.05 -0.116467213 44.00 -122.91 0.141102413 45.09 -123.04 -0.078632692 45.29 -122.97 -0.192394823 45.34 -122.64 -0.031490892 45.34 -122.65 -0.100540579 45.35 -122.68 0.080552588 45.37 -122.60 -0.110011813 45.40 -122.56 -0.006710302 45.58 -122.76 -0.096704215

Table G-11. Estimated Concentrations of TSS based on the LME Model Equation (the range of THg Concentration values were based on summary results)

TSS Concentrations (mg/L) THg Concentrations (ng/L) 4.27e-14 0.140 8.14e-13 0.622 1.94e-12 0.964 5.01e-12 1.560 7.41e-12 1.901 1.36e-11 2.590 5.53e-11 5.260

Oregon Department of Environmental Quality 185 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Table G-12. Estimated Percent Reduction of THg Concentrations based on the LME Model TSS Concentration Reduction (%) THg Concentration Reduction (%) 20 (from 100 to 80 mg/L) 10.7 40 (from 100 to 60 mg/L) 22.8 60 (from 100 to 40 mg/L) 37.1 70 (from 100 to 30 mg/L) 45.7 80 (from 100 to 20 mg/L) 55.7 90 (from 100 to 10 mg/L) 68.9

Table G-13. Schedule for interim goals for reducing TSS concentrations Years from Reduction in 75 Estimated Reductions in Cumulative Reduction in 2019 Percentile of TSS TSS Concentration (mg/L) THg Concentration (ng/L) 0 0% 17a 0 5 10% 15 2 10 25% 13 4 20 50% 8.5 8.5 30 75% 4.2 13 Note: a Based on the 75 Percentile of TSS Concentrations from the Dataset of 63 Surrogate Pairs. The current TSS Estimate Concentration is 17 mg/L.

Oregon Department of Environmental Quality 186 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

1.7 Figures

Figure G-1. Boxplot of Untransformed TSS (mg/L) and THg Concentrations (ng/L) from the Original Dataset of 187 Surrogate Pairs. The Red line represents an Instream Target Mercury Concentration of 0.14 ng/L.

Oregon Department of Environmental Quality 187 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Figure G-2. Dry and Wet Season Scatterplot of Untransformed TSS (mg/L) and THg Concentrations (ng/L) from the Original Dataset of 187 Surrogate Pairs.

Oregon Department of Environmental Quality 188 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Figure G-3. Willamette River Basin Instream Surrogate TSS-THg Sampling Sites Map.

Oregon Department of Environmental Quality 189 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Figure G-4. Boxplot of Untransformed TSS (mg/L) and THg Concentrations (ng/L) based on the Dataset of 63 Surrogate Pairs. The Red line represents an Instream Target Mercury Concentration of 0.14 ng/L.

Oregon Department of Environmental Quality 190 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Figure G-5. Scatterplot of Untransformed TSS (mg/L) and THg Concentrations (ng/L) according to HUC 8 Subbasin based on the Dataset of 63 Surrogate Pairs.

Figure G-6. Scatterplot of TSS (mg/L) and THg Concentrations (ng/L) for each individual HUC 8 Subbasin based on the Dataset of 63 Surrogate Pairs.

Oregon Department of Environmental Quality 191 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Figure G-7. Dry and Wet Season Scatterplot of Untransformed TSS (mg/L) and THg Concentrations (ng/L) based on the Dataset of 63 Surrogate Pairs.

Figure G-8. Box-cox Transformation of Dependent Variable of THg concentrations

Oregon Department of Environmental Quality 192 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Figure G-9. OLS Model Scatterplot for all HUC 8 Subbasins based on the Dataset of 63 Surrogate Pairs.

Figure G-10. 95 Percent Prediction Interval (fixed effect only) for LME Model Regression with Seasons as an Additional Fixed Effect and with Sites as a Random Effect for the 63 Instream THg- TSS surrogate sample pairs. THg concentrations are ng/L and TSS Concentrations are mg/L.

Oregon Department of Environmental Quality 193 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Figure G-11. 95 Percent Prediction Interval (fixed effect only) for LME Model Regression with Sites as a Random Effect for the 63 Instream THg-TSS surrogate sample pairs. THg concentrations are ng/L and TSS Concentrations are mg/L.

Oregon Department of Environmental Quality 194 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Figure G-12. Diagnostic Plot for OLS Model for all 63 Instream THg-TSS Surrogate Samples. The residuals versus fitted values plot checks for the assumptions of linearity.

The Q-Q plot checks for the assumptions of normality, while the scale-location plot checks for homoscedasticity (equal variance). The Cook’s Distance plot (residuals versus leverage) helps identify any significant outliers that can influence the regression coefficients of the OLS model.

Oregon Department of Environmental Quality 195 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Figure G-13. Diagnostic Plot for LME Model with Seasons as an Additional Fixed Effect and Sites a Random Effect for all 63 Instream THg-TSS Surrogate Samples.

The residuals versus fitted values plot checks for the assumptions of linearity. The Q-Q plot checks for the assumptions of normality, while the scale-location plot checks for homoscedasticity (equal variance). The Cook’s Distance plot (residuals versus leverage) helps identify any significant outliers that can influence the regression coefficients of the LME model.

Oregon Department of Environmental Quality 196 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Figure G-14. Diagnostic Plot for LME Model with Sites as a Random Effect for all 63 Instream THg- TSS Surrogate Samples.

The residuals versus fitted values plot checks for the assumptions of linearity. The Q-Q plot checks for the assumptions of normality, while the scale-location plot checks for homoscedasticity (equal variance). The Cook’s Distance plot (residuals versus leverage) helps identify any significant outliers that can influence the regression coefficients of the LME model.

Oregon Department of Environmental Quality 197 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Figure G-15. Empirical Cumulative Distribution of TSS Concentrations across all sites.

Oregon Department of Environmental Quality 198 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Figure G-16. Empirical Cumulative Distribution of TSS Concentrations with Reductions in the Upper Quartile Range (75th percentile) of TSS Concentrations across all sites.

The 0% Reduction line represents the cdf plot of current TSS concentrations (mg/L) that were accounted for within the study (same as Figure G-15).

Oregon Department of Environmental Quality 199 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

2. References Conroy, M. J., & Peterson, J. T. (2013). Decision Making in Natural Resource Management: A Structured, Adaptive Approach. Hoboken: Wiley-Blackwell. Eckley, C. S., & Branfireun, B. (2008). Mercury mobilization in urban stormwater runoff. Science of the Total Environment, 403(1-3), 164-177. Eckley, C. S., & Branfireun, B. (2009). Simulated rain events on an urban roadway to understand the dynamics of mercury mobilization in stormwater runoff. Water research, 43(15), 3635-3646. Environmental Protection Agency. (2002). Method 1631, Revision E: Mercury in Water by Oxidation, Purge and Trap, and Cold Vapor Atomic Fluorescence Spectrometry. Retrieved from https://www.epa.gov/sites/production/files/2015-08/documents/method_1631e_2002.pdf Hajduk. G.K. (2017). Introduction to linear mixed models. Retrieved from https://ourcodingclub.github.io/2017/03/15/mixed-models.html Helsel, D. R. (1990). Less than obvious-statistical treatment of data below the detection limit. Environmental Science & Technology, 24(12), 1766-1774. Helsel, D. R. (2005). Nondetects and data analysis. Statistics for censored environmental data. Wiley-Interscience. Helwig, N.E. (2017). Linear-Mixed Effects Regression. Retrieved from http://users.stat.umn.edu/~helwig/notes/lmer-Notes.pdf R Core Team (2018). R: A language and environment for statistical computing. R Foundation for Statistical. Computing, Vienna, Austria. URL https://www.R-project.org/. Hope, B. K. (2006). An assessment of anthropogenic source impacts on mercury cycling in the Willamette Basin, Oregon, USA. Science of the total environment, 356(1-3), 165-191. Hope, B. K., & Rubin, J. R. (2005). Mercury levels and relationships in water, sediment, and fish tissue in the Willamette Basin, Oregon. Archives of environmental contamination and toxicology, 48(3), 367-380. Interstate Technology and Regulatory Council (ITRC). Groundwater Statistics and Monitoring Compliance: Statistical Tools for the Project Life Cycle. Retrieved from https://www.itrcweb.org/gsmc-1/Content/Resources/GSMCPDF.pdf Johnson, A. (2007). Quality Assurance Project Plan: Yakima River Chlorinated Pesticides, PCBs, Suspended Sediment, and Turbidity Total Maximum Daily Load Study State of Oregon: Oregon Department of Environmental Quality. (2002). Upper Klamath Lake Drainage TMDL and Water Quality Management Plan (WQMP) State of Oregon: Oregon Department of Environmental Quality. (2006). Willamette Basin Mercury TMDL State of Oregon: Oregon Department of Environmental Quality. (2019). Revised Willamette Basin Mercury TMDL: Draft for Public Comment TetraTech.(2018). Draft Memorandum: Potential THg Surrogate Measures. TetraTech. (2019). Mercury TMDL Development for the Willamette River Basin (Oregon) – Technical Support Document (PUBLIC REVIEW DRAFT).

Oregon Department of Environmental Quality 200 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

3. R Packages Used

captioner: Letaw Alathea (2015). captioner: Numbers Figures and Creates Simple Captions. R package version 2.2.3. https://CRAN.R-project.org/package=captioner dplyr: Hadley Wickham, Romain François, Lionel Henry and Kirill Müller (2018). dplyr: A Grammar of Data Manipulation. R package version 0.7.6. https://CRAN.R- project.org/package=dplyr ggplot2: H. Wickham. ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag New York, 2016. ggpubr: Alboukadel Kassambara (2018). ggpubr: ‘ggplot2’ Based Publication Ready Plots. R package version 0.2. https://CRAN.R-project.org/package=ggpubr jtools: Long JA (2018). jtools: Analysis and Presentation of Social Scientific Data. R package version 1.1.0, . knitr: Yihui Xie (2018). knitr: A General-Purpose Package for Dynamic Report Generation in R. R package version 1.21. latexpdf: Tim Bergsma (2018). latexpdf: Convert Tables to PDF or PNG. R package version 0.1.6.https://CRAN.R-project.org/package=latexpdf lme4: Douglas Bates, Martin Maechler, Ben Bolker, Steve Walker (2015). Fitting Linear Mixed- Effects Models Using lme4. Journal of Statistical Software, 67(1), 1-48. doi:10.18637/jss.v067.i01. MuMIn: Kamil Barton (2018). MuMIn: Multi-Model Inference. R package version 1.42.1. https://CRAN.R-project.org/package=MuMIn rmarkdown: JJ Allaire, Yihui Xie, Jonathan McPherson, Javier Luraschi, Kevin Ushey, Aron Atkins, Hadley Wickham, Joe Cheng and Winston Chang (2018). rmarkdown: Dynamic Documents for R. R package version 1.10. https://CRAN.R-project.org/package=rmarkdown rsq: Dabao Zhang (2018). rsq: R-Squared and Related Measures. R package version 1.1. https://CRAN.R-project.org/package=rsq SSN: Ver Hoef, J. M. and Peterson, E. E. (2010) A moving average approach for spatial statistical models of stream networks (with discussion). Journal of the American Statistical Association 105:6-18. DOI: 10.1198/jasa.2009.ap08248 tidyverse: Hadley Wickham (2017). tidyverse: Easily Install and Load the ‘Tidyverse’. R package version 1.2.1. https://CRAN.R-project.org/package=tidyverse yaml: Jeremy Stephens, Kirill Simonov, Yihui Xie, Zhuoer Dong, Hadley Wickham, Jeffrey Horner, reikoch, Will Beasley, BrendanO’Connor and Gregory R. Warnes (2018). yaml: Methods to Convert R Data to YAML and Back. R package version 2.2.0. https://CRAN.R- project.org/package=yaml

Oregon Department of Environmental Quality 201 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 Appendix H. Watershed-Based Plan Crosswalk

Congress enacted Section 319 of the Clean Water Act in 1987, which established a national program to control nonpoint sources of water pollution. Under section 319, states, territories and tribes receive grant money that supports a wide variety of activities including technical assistance, financial assistance, education, training, and monitoring to assess the success of specific nonpoint source implementation projects.

Through Oregon’s annual 319 funding agreement, EPA requires the Oregon Department of Environmental Quality to ensure a watershed based plan or acceptable alternative plan is in place before any 319 project funds may be approved for projects in that watershed. A watershed based plan, as presented in the 2003 319 grant guidelines, includes elements (a) through (i) presented below in Table 1. TMDL, WQMP and DMA Implementation Plans together may serve as watershed based plans to satisfy these requirements. Other watershed planning documents may also meet these objectives. For more information, see EPA’s “Handbook for Developing Watershed Plans to Restore and Protect Our Waters” published in March 2008, EPA 841-N-08-002.

For the Willamette Basin Mercury TMDL, DEQ anticipates that all nine elements of a watershed- based plan are addressed in either the TMDL or WQMP sections of this document, or will be addressed through DMA TMDL implementation plans following the issuance of the TMDL. DEQ will complete the checklist below (indicated by “TBD”) to verify DMA implementation plans, or other watershed planning documents satisfy the indicated watershed element. Once completed, DEQ may approve section 319 grants throughout the Willamette Basin that address erosion control and runoff prevention.

Table H-1. Watershed Based Plan Checklist for the Willamette Basin Mercury TMDL

Watershed Based Plan (9-Key Elements) Checklist for CWA 319 Grant Funding Hydrologic Unit Name/s (Hydrologic Unit Code/s): (1) Middle Fork Willamette River (17090001), (2) Coast Fork Willamette River (17090002), (3) Upper Willamette River (17090003), (4) McKenzie River (17090004), (5) North Santiam River (17090005), (6) South Santiam River (17090006), (7) Middle Willamette River (17090007), (8) Yamhill River (17090008), (9) Molalla-Pudding River (17090009), (10) Tualatin River (17090010), (11) Clackamas River (17090011), (12) Lower Willamette River (17090012)

Pollutants Addressed: Mercury and sediment through reduction of erosion and runoff. Temperature impairments may also be addressed if 319 projects, such as implementing riparian streambank erosion control measures also contribute to temperature improvements, such as stream shading.

Watershed Based Plan/s: Willamette Basin TMDL and WQMP (2019), individual DMA implementation plans, watershed action plans, etc.

Oregon Department of Environmental Quality 202 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Element (A) Identification of Pollutant Causes and Sources I. The plan identifies causes of impairment and pollutant sources or groups of similar sources that need to be controlled to achieve needed load reductions. Watershed Based Plan/s Chapter, Section, Table, etc. Page No.(s) Willamette Basin Mercury TMDL and Section 5.1: Mercury cycling in Pg. 17 WQMP (2019) the environment

Figure 6-5 Pg. 27

Section 6.1.4: Nonpoint source Pg. 30 input data development

Section 6.2: Current total load Pg. 32 estimation for Willamette Basin

Section 9.1: Air emissions Pg. 40

Section 9.2: Nonpoint Sources Pg. 41

Section 9.4.2: Stormwater Pg. 51 permits and mercury loads

Element (B) Pollutant Load Reduction Estimates I. The plan identifies an estimate of the load reductions expected from management strategies.

Watershed Based Plan/s Chapter, Section, Table, etc. Page No.(s) Willamette Basin Mercury TMDL and Section 7.0: Loading capacity Pg. 34 WQMP (2019) and excess load

Table 10-1 Pg. 56

Section 10.1: Load allocations Pg. 58 for nonpoint sources

Element (C) Best Management Practices I. The plan provides a description of the nonpoint source management strategies to achieve load reductions. Watershed Based Plan/s Chapter, Section, Table, etc. Page No.(s) Willamette Basin Mercury TMDL and Section 13.3 Pg. 73 WQMP (2019)

DMA Implementation Plans, watershed TBD TBD action plans, etc. II. The plan describes critical areas in which the management strategies will be needed.

Oregon Department of Environmental Quality 203 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Watershed Based Plan/s Chapter, Section, Table, etc. Page No.(s) Willamette Basin Mercury TMDL and Section 9: Source Assessment Pg. 40 WQMP (2019)

DMA Implementation Plans, watershed TBD TBD action plans, etc.

Element (D) Financial and Technical Assistance I: The plan estimates the amounts of technical and financial assistance needed, associated costs, and/or the sources and authorities that will be relied upon to implement this plan. Watershed Based Plan/s Chapter, Section, Table, etc. Page No.(s) Willamette Basin Mercury TMDL and Section 13.7: Costs and Pg. 128 WQMP (2019) funding

DMA Implementation Plans, watershed TBD TBD action plans, etc.

Element (E) Education and Outreach I: The plan includes an information and education component used to enhance public understanding of the project and encourage their early and continued participation in selecting, designing, and implementing the nonpoint source management strategies that will be implemented. Watershed Based Plan/s Chapter, Section, Table, etc. Page No.(s) DMA Implementation Plans, watershed TBD TBD action plans, etc.

Element (F) Implementation Plan Schedule I: The plan includes a schedule for implementing the nonpoint source management measures that is reasonably expeditious. Watershed Based Plan/s Chapter, Section, Table, etc. Page No.(s) Willamette Basin Mercury TMDL and Section 13.4: Timeline for Pg. 124 WQMP (2019) implementing management strategies

DMA Implementation Plans, watershed TBD TBD action plans, etc.

Element (G) Interim Milestones I: The plan includes a description of interim measurable milestones for determining whether nonpoint source management strategies or other control actions are being implemented. Watershed Based Plan/s Chapter, Section, Table, etc. Page No.(s) Willamette Basin Mercury TMDL and Section 13.4: Timeline for Pg. 124 WQMP (2019) implementing management strategies

DMA Implementation Plans, watershed TBD TBD action plans, etc.

Oregon Department of Environmental Quality 204 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Element (H) Implementation Plan Effectiveness I: The plan includes a set of criteria that can be used to determine whether loading reductions are being achieved over time and substantial progress is being made toward attaining water quality standards (e.g. water quality condition or implementation actions). Watershed Based Plan/s Chapter, Section, Table, etc. Page No.(s) Willamette Basin Mercury TMDL and Section 13.6: Monitoring and Pg. 127 WQMP (2019) evaluation (will be a stand- alone document)

DMA Implementation Plans, watershed TBD TBD action plans, etc.

Element (I) Monitoring and Assessment I: The plan includes a monitoring component to evaluate the effectiveness of the implementation efforts over time, measured against the criteria established above under element H to determine whether loading reductions are being achieved over time and substantial progress is being made toward attaining water quality standards. Watershed Based Plan/s Chapter, Section, Table, etc. Page No.(s) Willamette Basin Mercury TMDL and Section 13.6: Monitoring and Pg. 127 WQMP (2019) evaluation (will be a stand- alone document)

DMA Implementation Plans, watershed TBD TBD action plans, etc.

Oregon Department of Environmental Quality 205 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019 Appendix I. Select HUC8-Level Analysis of Existing Condition and Conservatism of At- Source vs. Delivered Loads

2012-2019 Data and Analysis of Total Mercury in the Tualatin and Lower Willamette HUC8s

To minimize the uncertainty due to differences between HUC8s in numbers of data observations, collection locations (e.g., tributaries vs mainstem, headwaters vs outlet) and other regional geographic variations, DEQ calculated existing condition as a basin-wide surface water total mercury median concentration from the observed total mercury concentrations within the dataset used for TMDL analyses. Due to these known data differentials, the median concentrations within the Tualatin and Lower Willamette HUC8s was slightly above the basin- wide existing condition median of 1.2 ng/L. Each HUC8 median is within the distribution of the majority of the data used to calculate the basin-wide median and therefore, need not be treated differently from the rest of the basin. However, DEQ evaluated additional data collected from the mainstems of the Tualatin and Lower Willamette Rivers and during 2012 through 2019, a period more recent than the datasets used in the TMDL (through 2011). To target data representative of listed reaches and HUC8 outlets, DEQ selected sampling locations that were on the mainstems and close to the center of the channel and excluded sample locations near the shores of the rivers. No data from the tributaries in the subbasins were included in this updated dataset because data from side channels and near the influence of tributaries is not representative of the listed reach or the outlet of the HUC8. This updated dataset was not used in the TMDL modeling, but it improves the understanding of existing condition in these subbasins. This data will be a distinct dataset available on DEQ’s website as part of the TMDL datasets. A summary of the data characteristics and analysis results are provided in Table I-1 and maps of data collection locations are provided in Figure I-1 and Figure I-2.

Table I-1. General characteristics of recent total mercury data.

Difference between New Number of Number New THg THg and Data Data Collected Sampling of Time period Median Basin Median HUC8 Source By Locations Samples data covers (ng/L) (ng/L) (%) City of Portland 1.23 0.03 ng/L Lower DEQ- Bureau of Jan-2013 to 4 181 Willamette AWQMS Environmental Dec-2017 2% Services

Clean 1.32 0.12 ng/L Clean Water Mar-2012 to Tualatin Water 13 221 Services Oct-2019 Services 9%

Oregon Department of Environmental Quality 206 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Figure I-1. Sampling locations for more recent (2013-2017) Lower Willamette total mercury data. (Note: All locations are near the middle of the main channel of the Willamette River.)

Figure I-2. Sampling locations for more recent (2012-2019) Tualatin total mercury data. (Note: All locations are on the Tualatin River.)

Oregon Department of Environmental Quality 207 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Analysis of Tualatin/Dairy Creek HUC8-level At-source vs. Delivered loads To represent existing condition in the Willamette Basin, DEQ opted to use the at-source loads of total mercury modeled to be delivered from the land to the waterways within the Basin, rather than using the lower delivered loads that also account for losses in transit to the outlet of the Basin at the Columbia River. While the delivered loads are the more accurate estimate of existing condition, using the 65 percent higher basin-wide at-source loads provides conservatism to help address inherent uncertainty in modeling and differences between geography and observed data quantities across the HUC8s which make up the Basin. This conservatism is expressed through the larger existing condition loads needing greater reductions to achieve the water column target.

DEQ worked with EPA and Tetra Tech to investigate how at-source loads are more conservative through an example simulation for the Tualatin HUC8. The Tualatin HUC8 was simulated as three watersheds in the HSPF model as shown in Figure I-3. Of these, only the Tualatin River is listed as impaired based on the fish consumption advisories due to level of methylmercury in fish tissue. The losses in mercury during transit are considered a part of the overall reduction for mercury sources in tributaries of the impaired segments. To demonstrate how these losses can be explicitly accounted for from the Mass Balance Model output, DEQ requested that changes be made to the HSPF model to represent reduction in soil erosion within the Dairy Creek watershed. For this example, an 88 percent reduction in erosion occurring on forest, shrub and agricultural lands in the Dairy Creek watershed was simulated in the HSPF model (referred to as Scenario 1). The remaining sources through the Dairy Creek watershed and the rest of the Tualatin subbasin remain the same. The simulated total mercury at-source and delivered loads in Dairy Creek are listed in Table I-2.

Table I-2. Annual Average sediment associated loads for Dairy Creek

Scenario At-Source (kg/yr) Delivered (kg/yr Existing Condition 4.13 3.04 Scenario 1 2.44 1.81 Source: Tables 1 and 2 from (Tetra Tech, 2019b)

DEQ treats the loads from Dairy Creek as sources to the Tualatin River the same as any land area sources directly discharging into the Tualatin River. Therefore, any losses during transit from land area sources in the Dairy Creek watershed to the outlet of the Tualatin River are additional reductions for the Dairy Creek sources. This is captured in the following equation:

% = % + %

𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂 𝐷𝐷𝐷𝐷𝑖𝑖𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟 The percent loss 𝑅𝑅for𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 the Dairy𝑅𝑅 Creek sources 𝐿𝐿to𝐿𝐿𝐿𝐿𝐿𝐿 the outlet𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 of the𝑅𝑅 Tualatin River was quantified using the information provided by Tetra Tech. First, the load loss during transport from Dairy Creek to the outlet of the Tualatin River was calculated as:

=

𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 For Dairy Creek,𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿 the at-source load𝐿𝐿𝐿𝐿 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿reduction is𝐿𝐿𝐿𝐿 the difference− 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿 between 𝐿𝐿the𝐿𝐿 at-source existing condition and Scenario 1 from Table I-2:

Oregon Department of Environmental Quality 208 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

= = 4.13 2.44 𝑘𝑘𝑘𝑘 𝑘𝑘𝑘𝑘 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸 − 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿=𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆1.69 1 𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴 − 𝐿𝐿 𝐿𝐿 𝑘𝑘𝑘𝑘 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴 The delivered load reduction for Dairy Creek is the difference between𝐿𝐿 the existing condition and Scenario 1 delivered loads from Table I-2:

= = 3.04 1.81 𝑘𝑘𝑘𝑘 𝑘𝑘𝑘𝑘 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 − 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆= 1.231𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 − 𝐿𝐿 𝐿𝐿 𝑘𝑘𝑘𝑘 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 The load loss during transport from Dairy Creek is: 𝐿𝐿

= = 1.69 1.23 𝑘𝑘𝑘𝑘 𝑘𝑘𝑘𝑘 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴 − 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿= 0.46 𝐿𝐿 𝐿𝐿𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 − 𝐿𝐿 𝐿𝐿 𝑘𝑘𝑘𝑘 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 The percent loss for loads originating in Dairy Creek during transport𝐿𝐿 to the outlet of the Tualatin River are the percent of the total load delivered to the outlet of the Tualatin for Scenario 1:

% = 1 × 100 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 − 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿 � − � 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 1.23 0.46 % = 1 × 100 = 37% 𝑘𝑘𝑘𝑘 𝑘𝑘𝑘𝑘 1.23− 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿 � − 𝐿𝐿 𝐿𝐿 � 𝑘𝑘𝑘𝑘 𝐿𝐿 Re-arranging the overall percent reduction for Dairy Creek equation to get the percent reduction needed for at-source loads only in Dairy Creek:

% = % + %

𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 %𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 =𝐿𝐿𝐿𝐿%𝐿𝐿𝐿𝐿 𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 %

𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂 𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅% 𝑅𝑅 𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅= 88%𝑅𝑅 37% = 51%− 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿

𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷 By assigning the general𝑅𝑅 nonpoint𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑜𝑜𝑜𝑜 source load reduction of −88 percent to Dairy Creek, implementation of the TMDL will achieve a greater reduction than needed for the listed segments of the Tualatin River, and is therefore conservative in the Dairy Creek portion of the Tualatin HUC8 by an additional 37 percent.

Oregon Department of Environmental Quality 209 - 221 Final Revised Willamette Basin Mercury TMDL and WQMP November 2019

Figure I-3. Watersheds simulated in HSPF for the Tualatin HUC08 (17090010).

Oregon Department of Environmental Quality 210 - 221